{
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "l4xnhqYTKaEK"
      },
      "source": [
        "# Dynamics of Disease Transmission and Human Behavior Project - Time Series Analysis\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Obg9u9ZrnEQG"
      },
      "source": [
        "\n",
        "## 1. Goal\n",
        "\n",
        "* Research question: how well can be predict the number of COVID-19 cases one day in advance or one week in advance?\n",
        "* Math question: given the historical series of COVID-19 cases, what is the best estimator of the cases one day ahead or one week ahead? What is the test-sample mean squared error (MSE) of each estimate?\n",
        "\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Fpgfm6HknHBJ"
      },
      "source": [
        "## 2. Dataset\n",
        "\n",
        "The dataset contains confirmed COVID-19 cases for 50 states. The COVID-19 cases are reported on the John Hopkins platform. "
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 1,
      "metadata": {
        "id": "aIb9NfsfL4Vg"
      },
      "outputs": [],
      "source": [
        "import pandas as pd\n",
        "import numpy as np\n",
        "import matplotlib.pyplot as plt\n",
        "from sklearn.metrics import mean_squared_error\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 2,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ZLueIIW_J5hf",
        "outputId": "ab4f733c-5dfb-4eda-ec6b-3a6df7f72989"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Drive already mounted at /content/drive; to attempt to forcibly remount, call drive.mount(\"/content/drive\", force_remount=True).\n"
          ]
        }
      ],
      "source": [
        "from google.colab import drive\n",
        "drive.mount(\"/content/drive\")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 3,
      "metadata": {
        "id": "8fPMjUcIKGfY"
      },
      "outputs": [],
      "source": [
        "california = pd.read_csv(\"./drive/MyDrive/CS109 Project/state_level_data/California.csv\",parse_dates=['date'],index_col=['date'])\n",
        "new_york = pd.read_csv(\"./drive/MyDrive/CS109 Project/state_level_data/New York.csv\",parse_dates=['date'],index_col=['date'])\n",
        "ma = pd.read_csv(\"./drive/MyDrive/CS109 Project/state_level_data/Massachusetts.csv\",parse_dates=['date'],index_col=['date'])"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 4,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 487
        },
        "id": "sfQlKmIrPRVc",
        "outputId": "088b7bed-5055-4a7f-850f-4da286640a6c"
      },
      "outputs": [
        {
          "data": {
            "text/html": [
              "\n",
              "  <div id=\"df-7cbbf9ca-761c-4d7a-b439-880e3c7ac7a9\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>JHU_cases</th>\n",
              "      <th>JHU_deaths</th>\n",
              "      <th>JHU_hospitalizations</th>\n",
              "      <th>up2date</th>\n",
              "      <th>gt_after covid vaccine</th>\n",
              "      <th>gt_side effects of vaccine</th>\n",
              "      <th>gt_effects of covid vaccine</th>\n",
              "      <th>gt_covid</th>\n",
              "      <th>gt_how long does covid last</th>\n",
              "      <th>gt_anosmia</th>\n",
              "      <th>...</th>\n",
              "      <th>neighbor_South Dakota</th>\n",
              "      <th>neighbor_Tennessee</th>\n",
              "      <th>neighbor_Texas</th>\n",
              "      <th>neighbor_Utah</th>\n",
              "      <th>neighbor_Vermont</th>\n",
              "      <th>neighbor_Virginia</th>\n",
              "      <th>neighbor_Washington</th>\n",
              "      <th>neighbor_West Virginia</th>\n",
              "      <th>neighbor_Wisconsin</th>\n",
              "      <th>neighbor_Wyoming</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>date</th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>2020-01-01</th>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>...</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2020-01-02</th>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>...</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2020-01-03</th>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>...</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2020-01-04</th>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>108.406849</td>\n",
              "      <td>...</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2020-01-05</th>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>...</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>5 rows × 498 columns</p>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-7cbbf9ca-761c-4d7a-b439-880e3c7ac7a9')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-7cbbf9ca-761c-4d7a-b439-880e3c7ac7a9 button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-7cbbf9ca-761c-4d7a-b439-880e3c7ac7a9');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ],
            "text/plain": [
              "            JHU_cases  JHU_deaths  JHU_hospitalizations  up2date  \\\n",
              "date                                                               \n",
              "2020-01-01        NaN         NaN                   NaN      NaN   \n",
              "2020-01-02        NaN         NaN                   NaN      NaN   \n",
              "2020-01-03        NaN         NaN                   NaN      NaN   \n",
              "2020-01-04        NaN         NaN                   NaN      NaN   \n",
              "2020-01-05        NaN         NaN                   NaN      NaN   \n",
              "\n",
              "            gt_after covid vaccine  gt_side effects of vaccine  \\\n",
              "date                                                             \n",
              "2020-01-01                     0.0                         0.0   \n",
              "2020-01-02                     0.0                         0.0   \n",
              "2020-01-03                     0.0                         0.0   \n",
              "2020-01-04                     0.0                         0.0   \n",
              "2020-01-05                     0.0                         0.0   \n",
              "\n",
              "            gt_effects of covid vaccine  gt_covid  \\\n",
              "date                                                \n",
              "2020-01-01                          0.0       0.0   \n",
              "2020-01-02                          0.0       0.0   \n",
              "2020-01-03                          0.0       0.0   \n",
              "2020-01-04                          0.0       0.0   \n",
              "2020-01-05                          0.0       0.0   \n",
              "\n",
              "            gt_how long does covid last  gt_anosmia  ...  \\\n",
              "date                                                 ...   \n",
              "2020-01-01                          0.0    0.000000  ...   \n",
              "2020-01-02                          0.0    0.000000  ...   \n",
              "2020-01-03                          0.0    0.000000  ...   \n",
              "2020-01-04                          0.0  108.406849  ...   \n",
              "2020-01-05                          0.0    0.000000  ...   \n",
              "\n",
              "            neighbor_South Dakota  neighbor_Tennessee  neighbor_Texas  \\\n",
              "date                                                                    \n",
              "2020-01-01                    NaN                 NaN             NaN   \n",
              "2020-01-02                    NaN                 NaN             NaN   \n",
              "2020-01-03                    NaN                 NaN             NaN   \n",
              "2020-01-04                    NaN                 NaN             NaN   \n",
              "2020-01-05                    NaN                 NaN             NaN   \n",
              "\n",
              "            neighbor_Utah  neighbor_Vermont  neighbor_Virginia  \\\n",
              "date                                                             \n",
              "2020-01-01            NaN               NaN                NaN   \n",
              "2020-01-02            NaN               NaN                NaN   \n",
              "2020-01-03            NaN               NaN                NaN   \n",
              "2020-01-04            NaN               NaN                NaN   \n",
              "2020-01-05            NaN               NaN                NaN   \n",
              "\n",
              "            neighbor_Washington  neighbor_West Virginia  neighbor_Wisconsin  \\\n",
              "date                                                                          \n",
              "2020-01-01                  NaN                     NaN                 NaN   \n",
              "2020-01-02                  NaN                     NaN                 NaN   \n",
              "2020-01-03                  NaN                     NaN                 NaN   \n",
              "2020-01-04                  NaN                     NaN                 NaN   \n",
              "2020-01-05                  NaN                     NaN                 NaN   \n",
              "\n",
              "            neighbor_Wyoming  \n",
              "date                          \n",
              "2020-01-01               NaN  \n",
              "2020-01-02               NaN  \n",
              "2020-01-03               NaN  \n",
              "2020-01-04               NaN  \n",
              "2020-01-05               NaN  \n",
              "\n",
              "[5 rows x 498 columns]"
            ]
          },
          "execution_count": 4,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "california.head()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 5,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 470
        },
        "id": "1ZJKveACUW_d",
        "outputId": "248bde2c-2928-4b7e-cbf9-df3b1e8098c3"
      },
      "outputs": [
        {
          "data": {
            "text/html": [
              "\n",
              "  <div id=\"df-37b8b65d-82a2-4164-bfd4-14269d0499b8\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>JHU_cases</th>\n",
              "      <th>JHU_deaths</th>\n",
              "      <th>JHU_hospitalizations</th>\n",
              "      <th>up2date</th>\n",
              "      <th>gt_after covid vaccine</th>\n",
              "      <th>gt_side effects of vaccine</th>\n",
              "      <th>gt_effects of covid vaccine</th>\n",
              "      <th>gt_covid</th>\n",
              "      <th>gt_how long does covid last</th>\n",
              "      <th>gt_anosmia</th>\n",
              "      <th>...</th>\n",
              "      <th>neighbor_South Dakota</th>\n",
              "      <th>neighbor_Tennessee</th>\n",
              "      <th>neighbor_Texas</th>\n",
              "      <th>neighbor_Utah</th>\n",
              "      <th>neighbor_Vermont</th>\n",
              "      <th>neighbor_Virginia</th>\n",
              "      <th>neighbor_Washington</th>\n",
              "      <th>neighbor_West Virginia</th>\n",
              "      <th>neighbor_Wisconsin</th>\n",
              "      <th>neighbor_Wyoming</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>date</th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>2022-01-09</th>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>1036.271881</td>\n",
              "      <td>79.980083</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>191709.747309</td>\n",
              "      <td>880.874506</td>\n",
              "      <td>0.0</td>\n",
              "      <td>...</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2022-01-10</th>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>501.666380</td>\n",
              "      <td>359.532743</td>\n",
              "      <td>215.179826</td>\n",
              "      <td>222023.793840</td>\n",
              "      <td>719.959308</td>\n",
              "      <td>0.0</td>\n",
              "      <td>...</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2022-01-11</th>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>737.261085</td>\n",
              "      <td>221.918779</td>\n",
              "      <td>73.787815</td>\n",
              "      <td>224041.563172</td>\n",
              "      <td>1185.038030</td>\n",
              "      <td>0.0</td>\n",
              "      <td>...</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2022-01-12</th>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>978.821343</td>\n",
              "      <td>151.092032</td>\n",
              "      <td>150.713937</td>\n",
              "      <td>204037.539213</td>\n",
              "      <td>453.839532</td>\n",
              "      <td>0.0</td>\n",
              "      <td>...</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2022-01-13</th>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>984.362561</td>\n",
              "      <td>75.973690</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>189624.711109</td>\n",
              "      <td>1141.021922</td>\n",
              "      <td>0.0</td>\n",
              "      <td>...</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>5 rows × 498 columns</p>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-37b8b65d-82a2-4164-bfd4-14269d0499b8')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-37b8b65d-82a2-4164-bfd4-14269d0499b8 button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-37b8b65d-82a2-4164-bfd4-14269d0499b8');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ],
            "text/plain": [
              "            JHU_cases  JHU_deaths  JHU_hospitalizations  up2date  \\\n",
              "date                                                               \n",
              "2022-01-09        NaN         NaN                   NaN      NaN   \n",
              "2022-01-10        NaN         NaN                   NaN      NaN   \n",
              "2022-01-11        NaN         NaN                   NaN      NaN   \n",
              "2022-01-12        NaN         NaN                   NaN      NaN   \n",
              "2022-01-13        NaN         NaN                   NaN      NaN   \n",
              "\n",
              "            gt_after covid vaccine  gt_side effects of vaccine  \\\n",
              "date                                                             \n",
              "2022-01-09             1036.271881                   79.980083   \n",
              "2022-01-10              501.666380                  359.532743   \n",
              "2022-01-11              737.261085                  221.918779   \n",
              "2022-01-12              978.821343                  151.092032   \n",
              "2022-01-13              984.362561                   75.973690   \n",
              "\n",
              "            gt_effects of covid vaccine       gt_covid  \\\n",
              "date                                                     \n",
              "2022-01-09                     0.000000  191709.747309   \n",
              "2022-01-10                   215.179826  222023.793840   \n",
              "2022-01-11                    73.787815  224041.563172   \n",
              "2022-01-12                   150.713937  204037.539213   \n",
              "2022-01-13                     0.000000  189624.711109   \n",
              "\n",
              "            gt_how long does covid last  gt_anosmia  ...  \\\n",
              "date                                                 ...   \n",
              "2022-01-09                   880.874506         0.0  ...   \n",
              "2022-01-10                   719.959308         0.0  ...   \n",
              "2022-01-11                  1185.038030         0.0  ...   \n",
              "2022-01-12                   453.839532         0.0  ...   \n",
              "2022-01-13                  1141.021922         0.0  ...   \n",
              "\n",
              "            neighbor_South Dakota  neighbor_Tennessee  neighbor_Texas  \\\n",
              "date                                                                    \n",
              "2022-01-09                    NaN                 NaN             NaN   \n",
              "2022-01-10                    NaN                 NaN             NaN   \n",
              "2022-01-11                    NaN                 NaN             NaN   \n",
              "2022-01-12                    NaN                 NaN             NaN   \n",
              "2022-01-13                    NaN                 NaN             NaN   \n",
              "\n",
              "            neighbor_Utah  neighbor_Vermont  neighbor_Virginia  \\\n",
              "date                                                             \n",
              "2022-01-09            NaN               NaN                NaN   \n",
              "2022-01-10            NaN               NaN                NaN   \n",
              "2022-01-11            NaN               NaN                NaN   \n",
              "2022-01-12            NaN               NaN                NaN   \n",
              "2022-01-13            NaN               NaN                NaN   \n",
              "\n",
              "            neighbor_Washington  neighbor_West Virginia  neighbor_Wisconsin  \\\n",
              "date                                                                          \n",
              "2022-01-09                  NaN                     NaN                 NaN   \n",
              "2022-01-10                  NaN                     NaN                 NaN   \n",
              "2022-01-11                  NaN                     NaN                 NaN   \n",
              "2022-01-12                  NaN                     NaN                 NaN   \n",
              "2022-01-13                  NaN                     NaN                 NaN   \n",
              "\n",
              "            neighbor_Wyoming  \n",
              "date                          \n",
              "2022-01-09               NaN  \n",
              "2022-01-10               NaN  \n",
              "2022-01-11               NaN  \n",
              "2022-01-12               NaN  \n",
              "2022-01-13               NaN  \n",
              "\n",
              "[5 rows x 498 columns]"
            ]
          },
          "execution_count": 5,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "california.tail()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 6,
      "metadata": {
        "id": "Tedq77W4Ukv8"
      },
      "outputs": [],
      "source": [
        "# remove final 5 days because it does not make sense to have zero cases\n",
        "california = california.iloc[:-5, :]"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 7,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "e3CNjGyWPVcP",
        "outputId": "a9376123-5aaa-4435-8e8c-198cbab32384"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "<class 'pandas.core.frame.DataFrame'>\n",
            "DatetimeIndex: 739 entries, 2020-01-01 to 2022-01-08\n",
            "Columns: 498 entries, JHU_cases to neighbor_Wyoming\n",
            "dtypes: float64(498)\n",
            "memory usage: 2.8 MB\n"
          ]
        }
      ],
      "source": [
        "california.info()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 8,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 446
        },
        "id": "-5TyFpaNPY7a",
        "outputId": "8ed0dda1-a06e-4e20-e7ca-e08021b3253a"
      },
      "outputs": [
        {
          "data": {
            "text/html": [
              "\n",
              "  <div id=\"df-14986c9e-c928-450c-b9d0-669a9c9255b3\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>JHU_cases</th>\n",
              "      <th>JHU_deaths</th>\n",
              "      <th>JHU_hospitalizations</th>\n",
              "      <th>up2date</th>\n",
              "      <th>gt_after covid vaccine</th>\n",
              "      <th>gt_side effects of vaccine</th>\n",
              "      <th>gt_effects of covid vaccine</th>\n",
              "      <th>gt_covid</th>\n",
              "      <th>gt_how long does covid last</th>\n",
              "      <th>gt_anosmia</th>\n",
              "      <th>...</th>\n",
              "      <th>neighbor_South Dakota</th>\n",
              "      <th>neighbor_Tennessee</th>\n",
              "      <th>neighbor_Texas</th>\n",
              "      <th>neighbor_Utah</th>\n",
              "      <th>neighbor_Vermont</th>\n",
              "      <th>neighbor_Virginia</th>\n",
              "      <th>neighbor_Washington</th>\n",
              "      <th>neighbor_West Virginia</th>\n",
              "      <th>neighbor_Wisconsin</th>\n",
              "      <th>neighbor_Wyoming</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>count</th>\n",
              "      <td>718.000000</td>\n",
              "      <td>718.000000</td>\n",
              "      <td>544.000000</td>\n",
              "      <td>416.000000</td>\n",
              "      <td>739.000000</td>\n",
              "      <td>739.000000</td>\n",
              "      <td>739.000000</td>\n",
              "      <td>739.000000</td>\n",
              "      <td>739.000000</td>\n",
              "      <td>739.000000</td>\n",
              "      <td>...</td>\n",
              "      <td>718.000000</td>\n",
              "      <td>718.000000</td>\n",
              "      <td>718.000000</td>\n",
              "      <td>718.000000</td>\n",
              "      <td>718.000000</td>\n",
              "      <td>718.000000</td>\n",
              "      <td>718.000000</td>\n",
              "      <td>718.000000</td>\n",
              "      <td>718.000000</td>\n",
              "      <td>718.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>mean</th>\n",
              "      <td>8408.796657</td>\n",
              "      <td>107.302228</td>\n",
              "      <td>654.398897</td>\n",
              "      <td>266.588302</td>\n",
              "      <td>1031.942497</td>\n",
              "      <td>267.194675</td>\n",
              "      <td>185.924657</td>\n",
              "      <td>90028.915062</td>\n",
              "      <td>261.722963</td>\n",
              "      <td>29.050446</td>\n",
              "      <td>...</td>\n",
              "      <td>263.803621</td>\n",
              "      <td>2037.598886</td>\n",
              "      <td>6963.263231</td>\n",
              "      <td>949.910864</td>\n",
              "      <td>106.435933</td>\n",
              "      <td>1700.607242</td>\n",
              "      <td>1296.756267</td>\n",
              "      <td>495.643454</td>\n",
              "      <td>1651.902507</td>\n",
              "      <td>166.931755</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>std</th>\n",
              "      <td>12651.020881</td>\n",
              "      <td>138.156802</td>\n",
              "      <td>557.909209</td>\n",
              "      <td>194.928115</td>\n",
              "      <td>1390.723867</td>\n",
              "      <td>345.901801</td>\n",
              "      <td>257.298072</td>\n",
              "      <td>48033.048924</td>\n",
              "      <td>261.124650</td>\n",
              "      <td>55.944590</td>\n",
              "      <td>...</td>\n",
              "      <td>377.790566</td>\n",
              "      <td>3500.653803</td>\n",
              "      <td>9728.762811</td>\n",
              "      <td>1276.364174</td>\n",
              "      <td>238.298140</td>\n",
              "      <td>2672.792276</td>\n",
              "      <td>2221.208234</td>\n",
              "      <td>696.125703</td>\n",
              "      <td>2286.520080</td>\n",
              "      <td>248.876616</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>min</th>\n",
              "      <td>-3935.000000</td>\n",
              "      <td>-364.000000</td>\n",
              "      <td>116.000000</td>\n",
              "      <td>0.305869</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>...</td>\n",
              "      <td>-4.000000</td>\n",
              "      <td>-42.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>-1.000000</td>\n",
              "      <td>-232.000000</td>\n",
              "      <td>-39.000000</td>\n",
              "      <td>-4.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>-8.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>25%</th>\n",
              "      <td>1808.000000</td>\n",
              "      <td>25.000000</td>\n",
              "      <td>321.000000</td>\n",
              "      <td>148.166794</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>59592.925370</td>\n",
              "      <td>85.080935</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>...</td>\n",
              "      <td>8.000000</td>\n",
              "      <td>80.250000</td>\n",
              "      <td>1405.250000</td>\n",
              "      <td>164.000000</td>\n",
              "      <td>4.000000</td>\n",
              "      <td>338.500000</td>\n",
              "      <td>179.000000</td>\n",
              "      <td>37.250000</td>\n",
              "      <td>170.000000</td>\n",
              "      <td>4.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>50%</th>\n",
              "      <td>4261.500000</td>\n",
              "      <td>71.000000</td>\n",
              "      <td>432.500000</td>\n",
              "      <td>218.744994</td>\n",
              "      <td>452.973872</td>\n",
              "      <td>106.062503</td>\n",
              "      <td>76.018212</td>\n",
              "      <td>85254.949587</td>\n",
              "      <td>202.678864</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>...</td>\n",
              "      <td>103.000000</td>\n",
              "      <td>1138.500000</td>\n",
              "      <td>4353.500000</td>\n",
              "      <td>469.500000</td>\n",
              "      <td>28.500000</td>\n",
              "      <td>985.000000</td>\n",
              "      <td>662.000000</td>\n",
              "      <td>245.000000</td>\n",
              "      <td>722.500000</td>\n",
              "      <td>59.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>75%</th>\n",
              "      <td>9263.500000</td>\n",
              "      <td>126.000000</td>\n",
              "      <td>767.250000</td>\n",
              "      <td>372.607233</td>\n",
              "      <td>1545.638911</td>\n",
              "      <td>433.671560</td>\n",
              "      <td>303.376619</td>\n",
              "      <td>119532.223229</td>\n",
              "      <td>343.857587</td>\n",
              "      <td>74.962550</td>\n",
              "      <td>...</td>\n",
              "      <td>391.750000</td>\n",
              "      <td>2448.250000</td>\n",
              "      <td>9097.250000</td>\n",
              "      <td>1342.250000</td>\n",
              "      <td>133.000000</td>\n",
              "      <td>1947.500000</td>\n",
              "      <td>1644.750000</td>\n",
              "      <td>800.500000</td>\n",
              "      <td>2457.750000</td>\n",
              "      <td>214.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>max</th>\n",
              "      <td>133669.000000</td>\n",
              "      <td>1174.000000</td>\n",
              "      <td>2580.000000</td>\n",
              "      <td>1038.286656</td>\n",
              "      <td>7638.661166</td>\n",
              "      <td>2015.938064</td>\n",
              "      <td>1288.312759</td>\n",
              "      <td>279350.860227</td>\n",
              "      <td>1762.854735</td>\n",
              "      <td>489.629451</td>\n",
              "      <td>...</td>\n",
              "      <td>3047.000000</td>\n",
              "      <td>41464.000000</td>\n",
              "      <td>162871.000000</td>\n",
              "      <td>14754.000000</td>\n",
              "      <td>2779.000000</td>\n",
              "      <td>40246.000000</td>\n",
              "      <td>33069.000000</td>\n",
              "      <td>9164.000000</td>\n",
              "      <td>16956.000000</td>\n",
              "      <td>1541.000000</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>8 rows × 498 columns</p>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-14986c9e-c928-450c-b9d0-669a9c9255b3')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-14986c9e-c928-450c-b9d0-669a9c9255b3 button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-14986c9e-c928-450c-b9d0-669a9c9255b3');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ],
            "text/plain": [
              "           JHU_cases   JHU_deaths  JHU_hospitalizations      up2date  \\\n",
              "count     718.000000   718.000000            544.000000   416.000000   \n",
              "mean     8408.796657   107.302228            654.398897   266.588302   \n",
              "std     12651.020881   138.156802            557.909209   194.928115   \n",
              "min     -3935.000000  -364.000000            116.000000     0.305869   \n",
              "25%      1808.000000    25.000000            321.000000   148.166794   \n",
              "50%      4261.500000    71.000000            432.500000   218.744994   \n",
              "75%      9263.500000   126.000000            767.250000   372.607233   \n",
              "max    133669.000000  1174.000000           2580.000000  1038.286656   \n",
              "\n",
              "       gt_after covid vaccine  gt_side effects of vaccine  \\\n",
              "count              739.000000                  739.000000   \n",
              "mean              1031.942497                  267.194675   \n",
              "std               1390.723867                  345.901801   \n",
              "min                  0.000000                    0.000000   \n",
              "25%                  0.000000                    0.000000   \n",
              "50%                452.973872                  106.062503   \n",
              "75%               1545.638911                  433.671560   \n",
              "max               7638.661166                 2015.938064   \n",
              "\n",
              "       gt_effects of covid vaccine       gt_covid  \\\n",
              "count                   739.000000     739.000000   \n",
              "mean                    185.924657   90028.915062   \n",
              "std                     257.298072   48033.048924   \n",
              "min                       0.000000       0.000000   \n",
              "25%                       0.000000   59592.925370   \n",
              "50%                      76.018212   85254.949587   \n",
              "75%                     303.376619  119532.223229   \n",
              "max                    1288.312759  279350.860227   \n",
              "\n",
              "       gt_how long does covid last  gt_anosmia  ...  neighbor_South Dakota  \\\n",
              "count                   739.000000  739.000000  ...             718.000000   \n",
              "mean                    261.722963   29.050446  ...             263.803621   \n",
              "std                     261.124650   55.944590  ...             377.790566   \n",
              "min                       0.000000    0.000000  ...              -4.000000   \n",
              "25%                      85.080935    0.000000  ...               8.000000   \n",
              "50%                     202.678864    0.000000  ...             103.000000   \n",
              "75%                     343.857587   74.962550  ...             391.750000   \n",
              "max                    1762.854735  489.629451  ...            3047.000000   \n",
              "\n",
              "       neighbor_Tennessee  neighbor_Texas  neighbor_Utah  neighbor_Vermont  \\\n",
              "count          718.000000      718.000000     718.000000        718.000000   \n",
              "mean          2037.598886     6963.263231     949.910864        106.435933   \n",
              "std           3500.653803     9728.762811    1276.364174        238.298140   \n",
              "min            -42.000000        0.000000       0.000000         -1.000000   \n",
              "25%             80.250000     1405.250000     164.000000          4.000000   \n",
              "50%           1138.500000     4353.500000     469.500000         28.500000   \n",
              "75%           2448.250000     9097.250000    1342.250000        133.000000   \n",
              "max          41464.000000   162871.000000   14754.000000       2779.000000   \n",
              "\n",
              "       neighbor_Virginia  neighbor_Washington  neighbor_West Virginia  \\\n",
              "count         718.000000           718.000000              718.000000   \n",
              "mean         1700.607242          1296.756267              495.643454   \n",
              "std          2672.792276          2221.208234              696.125703   \n",
              "min          -232.000000           -39.000000               -4.000000   \n",
              "25%           338.500000           179.000000               37.250000   \n",
              "50%           985.000000           662.000000              245.000000   \n",
              "75%          1947.500000          1644.750000              800.500000   \n",
              "max         40246.000000         33069.000000             9164.000000   \n",
              "\n",
              "       neighbor_Wisconsin  neighbor_Wyoming  \n",
              "count          718.000000        718.000000  \n",
              "mean          1651.902507        166.931755  \n",
              "std           2286.520080        248.876616  \n",
              "min              0.000000         -8.000000  \n",
              "25%            170.000000          4.000000  \n",
              "50%            722.500000         59.000000  \n",
              "75%           2457.750000        214.000000  \n",
              "max          16956.000000       1541.000000  \n",
              "\n",
              "[8 rows x 498 columns]"
            ]
          },
          "execution_count": 8,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "california.describe()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 9,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 404
        },
        "id": "AYajleAHLgFC",
        "outputId": "e1f307d2-9815-44ca-83e6-3eada433e7f0"
      },
      "outputs": [
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 864x432 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          },
          "output_type": "display_data"
        }
      ],
      "source": [
        "fig,ax = plt.subplots(1,1, figsize = (12,6))\n",
        "plt.plot(california.index.values, california['JHU_cases'])\n",
        "plt.xlabel('Time')\n",
        "plt.ylabel('Cases')\n",
        "plt.title('COVID-19 cases in California over time')\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "hpsEn1ELNP9B"
      },
      "source": [
        "## 3. Data Cleaning\n",
        "\n",
        "We observed that most of NAs are at the beginning of the padamic, so we believe the number of cases are 0 or close to 0. Therefore, we decided to replace nan with 0 in cases. We also flipped the sign of the negative number of cases, because they are likely to be erroneous data."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 10,
      "metadata": {
        "id": "QXcY4Oz3NfX7"
      },
      "outputs": [],
      "source": [
        "california['JHU_cases'] = california['JHU_cases'].fillna(0)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 11,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "57DHtWFQNrA6",
        "outputId": "79eb450b-edbe-4f7b-e19d-325cc1392e12"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "date\n",
              "2020-03-29   -2019.0\n",
              "2021-06-29   -3935.0\n",
              "Name: JHU_cases, dtype: float64"
            ]
          },
          "execution_count": 11,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "california['JHU_cases'][california['JHU_cases'] <0]"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 12,
      "metadata": {
        "id": "mrnss7D2OEi7"
      },
      "outputs": [],
      "source": [
        "california['JHU_cases'][california['JHU_cases'] == -2019.0] = 2019.0\n",
        "california['JHU_cases'][california['JHU_cases'] == -3935.0] = 3935.0"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 13,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "x152zzubOMbX",
        "outputId": "18d028fb-3b13-4451-bc82-6ff3eb0e5b66"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "Series([], Name: JHU_cases, dtype: float64)"
            ]
          },
          "execution_count": 13,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "california['JHU_cases'][california['JHU_cases'] <0]"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "HWN92W_tMrzt"
      },
      "source": [
        "## 4. Data Transformation\n",
        "\n",
        "Since the variance is not constant overtime, it violates one of the assumptions of stationarity. We can corrrect it by rolling average or a log transformation of the data."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 14,
      "metadata": {
        "id": "AkXl_ufBPLqp"
      },
      "outputs": [],
      "source": [
        "# Applying rolling average\n",
        "window = 5\n",
        "california['JHU_cases'] = california['JHU_cases'].transform(lambda x: x.rolling(window).mean())"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 15,
      "metadata": {
        "id": "ztoQ81yFVT1Y"
      },
      "outputs": [],
      "source": [
        "california['JHU_cases'] = california['JHU_cases'].fillna(0)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 16,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 404
        },
        "id": "jAiHiKZsPlyF",
        "outputId": "acc1813d-1281-4ab4-d4c8-6a9ae784b381"
      },
      "outputs": [
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 864x432 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          },
          "output_type": "display_data"
        }
      ],
      "source": [
        "fig,ax = plt.subplots(1,1, figsize = (12,6))\n",
        "plt.plot(california.index.values, california['JHU_cases'])\n",
        "plt.xlabel('Time')\n",
        "plt.ylabel('Cases')\n",
        "plt.title('COVID-19 cases in California over time')\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 17,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 390
        },
        "id": "sgX_hz12M6yo",
        "outputId": "ef997a30-7f8d-489a-b77d-9ae84f714f68"
      },
      "outputs": [
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 864x432 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          },
          "output_type": "display_data"
        }
      ],
      "source": [
        "# Applying log transformation\n",
        "fig,ax = plt.subplots(1,1, figsize = (12,6))\n",
        "log_cases = np.log(california['JHU_cases']+1)\n",
        "plt.plot(california.index.values, log_cases)\n",
        "plt.title(\"log of COVID-19 cases in CA over time\")\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 18,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "2-_p0ag7Tnk9",
        "outputId": "16943b87-4175-4e28-d8a6-5f46e538c79a"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "False"
            ]
          },
          "execution_count": 18,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "any(california.index.isna())"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 19,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "c0XNKaSmToOp",
        "outputId": "a81e8633-2e6a-4c9c-b308-6a99671ecfb8"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "False"
            ]
          },
          "execution_count": 19,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "any(log_cases.isna())"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "uL067l4LQMGa"
      },
      "source": [
        "\n",
        "### Information Criteria\n",
        "\n",
        "Idea: find a balance between model fit and complexity. \n",
        "\n",
        "AIC (Akaike Information Criteria): $-2 * \\text{log-likelihood} + 2k$ \n",
        "\n",
        "BIC (Bayesian Information Criteria): $-2 * \\text{log-likelihood} + k \\log(n)$ \n",
        "\n",
        "where $k$ is the model complexity (degrees of freedom used up) and $n$ is the number of elements. \n",
        "\n",
        "We aim to select a model that minimizes both of these criteria; when they disagree, turn to another approach like out-of-sample validation. "
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 20,
      "metadata": {
        "id": "j7kVsX2vQWNz"
      },
      "outputs": [],
      "source": [
        "# Define functions to calculate AIC and BIC\n",
        "from scipy.stats import norm \n",
        "\n",
        "def evaluate_AIC(k, residuals):\n",
        "  \"\"\"\n",
        "  Finds the AIC given the number of parameters estimated and \n",
        "  the residuals of the model. Assumes residuals are distributed \n",
        "  Gaussian with unknown variance. \n",
        "  \"\"\"\n",
        "  standard_deviation = np.std(residuals)\n",
        "  log_likelihood = norm.logpdf(residuals, 0, scale=standard_deviation)\n",
        "  return 2 * k - 2 * np.sum(log_likelihood)\n",
        "\n",
        "def evaluate_BIC(k, residuals):\n",
        "  \"\"\"\n",
        "  Finds the AIC given the number of parameters estimated and \n",
        "  the residuals of the model. Assumes residuals are distributed \n",
        "  Gaussian with unknown variance. \n",
        "  \"\"\"\n",
        "  standard_deviation = np.std(residuals)\n",
        "  log_likelihood = norm.logpdf(residuals, 0, scale=standard_deviation)\n",
        "  return k * np.log(len(residuals)) - 2 * np.sum(log_likelihood)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "8ln2yFsiRJFq"
      },
      "source": [
        "## 5. Models\n",
        "\n",
        "\n",
        "\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "NCz7WPmpnP4S"
      },
      "source": [
        "### Linear Model"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 21,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 785
        },
        "id": "rWFmSnGlRGKR",
        "outputId": "0b53746c-71f5-4cb3-f7b4-a6bc9f7add31"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "[0.00702823]\n",
            "MSE with linear fit: 4.548069813129548\n",
            "AIC: 3218.556614436924\n",
            "BIC: 3223.161912357872\n"
          ]
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 864x864 with 2 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          },
          "output_type": "display_data"
        }
      ],
      "source": [
        "from sklearn import linear_model\n",
        "clf = linear_model.LinearRegression()\n",
        "index_len = len(california.index.values)\n",
        "index = np.linspace(1,index_len,index_len).reshape(-1,1)\n",
        "# index = california.index.values.astype('datetime64[D]').reshape(-1,1)\n",
        "\n",
        "clf.fit(index, log_cases)\n",
        "print(clf.coef_) \n",
        "linear_prediction = clf.predict(index)\n",
        "\n",
        "fig, axs = plt.subplots(2,1, figsize = (12,12))\n",
        "axs[0].plot(california.index.values, log_cases, label='log COVID-19 cases (original data)')\n",
        "axs[0].plot(california.index.values, linear_prediction, 'r', label='fitted line')\n",
        "axs[0].legend()\n",
        "axs[0].set_title(\"True Log Cases vs Fitted Linear Line\")\n",
        "linear_residuals = log_cases - linear_prediction\n",
        "axs[1].plot(california.index.values, linear_residuals, 'o')\n",
        "axs[1].set_title(\"Residuals: True Log Cases - Fitted Linear Line\")\n",
        "\n",
        "print(\"MSE with linear fit:\", np.mean((linear_residuals)**2))\n",
        "print(\"AIC:\", evaluate_AIC(1, linear_residuals))\n",
        "print(\"BIC:\", evaluate_BIC(1, linear_residuals))"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "0EAdvCqIZFqo"
      },
      "source": [
        "### Higher Order Model"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 22,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 785
        },
        "id": "L7e1svOtZJDs",
        "outputId": "49596167-c9b3-4553-b1d8-420832c4d8b7"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "[ 8.76760836e-02 -2.21421396e-04  1.68901696e-07]\n",
            "MSE with cubic fit: 0.979912988137383\n",
            "AIC: 2086.195734408407\n",
            "BIC: 2095.4063302503037\n"
          ]
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 864x864 with 2 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          },
          "output_type": "display_data"
        }
      ],
      "source": [
        "# After linear fit, it seems like a higher order model is needed\n",
        "from sklearn import linear_model\n",
        "clf = linear_model.LinearRegression()\n",
        "index_len = len(california.index.values)\n",
        "index = np.linspace(1,index_len,index_len).reshape(-1,1)\n",
        "\n",
        "new_x = np.hstack((index, index ** 2, index ** 3))\n",
        "clf.fit(new_x, log_cases)\n",
        "print(clf.coef_) \n",
        "cubic_prediction = clf.predict(new_x)\n",
        "\n",
        "fig, axs = plt.subplots(2,1, figsize=(12,12))\n",
        "axs[0].plot(california.index.values, log_cases, label='log COVID-19 cases (original data)')\n",
        "axs[0].plot(california.index.values, cubic_prediction, 'r', label='fitted line')\n",
        "axs[0].legend()\n",
        "axs[0].set_title(\"True Log Cases vs Fitted Cubic Line\")\n",
        "cubic_residuals = log_cases - cubic_prediction\n",
        "axs[1].plot(california.index.values, cubic_residuals, 'o')\n",
        "axs[1].set_title(\"Residuals: True Log Cases - Fitted Cubic Line\")\n",
        "print(\"MSE with cubic fit:\", np.mean((cubic_residuals)**2))\n",
        "print(\"AIC:\", evaluate_AIC(2, cubic_residuals))\n",
        "print(\"BIC:\", evaluate_BIC(2, cubic_residuals))"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "C-qrUVY8bimf"
      },
      "source": [
        "## 6. Examine ACF/PACF\n",
        "\n",
        "Let's look at the ACF/PACF of the residuals after removing the cubic trend. This tells us about existence of periodic components, or autoregressive or moving average trends that we need to account for. "
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 23,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 662
        },
        "id": "vaLv6CKpbhrY",
        "outputId": "6e691d09-fec4-4656-c21f-11542d760458"
      },
      "outputs": [
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/statsmodels/graphics/tsaplots.py:353: FutureWarning: The default method 'yw' can produce PACF values outside of the [-1,1] interval. After 0.13, the default will change tounadjusted Yule-Walker ('ywm'). You can use this method now by setting method='ywm'.\n",
            "  FutureWarning,\n"
          ]
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 864x720 with 2 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          },
          "output_type": "display_data"
        }
      ],
      "source": [
        "import statsmodels.api as sm\n",
        "fig, axs = plt.subplots(2,1, figsize = (12,10))\n",
        "sm.graphics.tsa.plot_acf(cubic_residuals, lags=30, ax = axs[0])\n",
        "sm.graphics.tsa.plot_pacf(cubic_residuals, lags=30, ax = axs[1])\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Lg8iVTtzcwKS"
      },
      "source": [
        "After taking a rolling average, taking a log transformation of the data, and removing a cubic trend, there's a pattern in the ACF/PACF plots. Strong first two terms in PACF, exponentially decaying ACF. This pattern is an evidence of an AR(1) component. It is important to model these components in order to forecase the cases in the future."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "So1oCPTTdWl5"
      },
      "source": [
        "## 7. ARIMA Models\n",
        "\n",
        "There are open source tools we can use to estimate AR/MA terms. Here, we use the ARIMA package to fit models to the data. \n",
        "\n",
        "Estimation goal: given an order $p, q$, find estimates of $\\phi$ and $\\theta$\n",
        "\n",
        "The general form for an AR model of order $p$:\n",
        "$$X_t = \\phi_1 X_{t-1} + \\phi_2 X_{t-2} \\dots + \\phi_p X_{t - p} + \\varepsilon_t$$\n",
        "\n",
        "And the form for an MA model of order $q$:\n",
        "$$X_t = \\varepsilon_t + \\theta_1 \\varepsilon_{t-1} + \\theta_2 \\varepsilon_{t-2} \\dots + \\theta_q \\varepsilon_{t-1}$$\n",
        "\n",
        "where $ɛ_t$ is mean zero white noise that's uncorrelated with any lagged or future values. "
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 24,
      "metadata": {
        "id": "U_SIdB4rec7Q"
      },
      "outputs": [],
      "source": [
        "from statsmodels.tsa.arima.model import ARIMA\n",
        "\n",
        "def grid_search_ARIMA(data, AR_range, MA_range, verbose=False):\n",
        "  min_aic = np.inf \n",
        "  min_bic = np.inf\n",
        "  min_aic_index = None\n",
        "  min_bic_index = None \n",
        "  aic_matrix = np.zeros((len(AR_range), len(MA_range)))\n",
        "  bic_matrix = np.zeros((len(AR_range), len(MA_range)))\n",
        "  for AR_order in AR_range:\n",
        "    for MA_order in MA_range:\n",
        "      arma = ARIMA(data, order=(AR_order, 0, MA_order)).fit()\n",
        "      aic_matrix[AR_order, MA_order] = arma.aic\n",
        "      bic_matrix[AR_order, MA_order] = arma.bic\n",
        "      if arma.aic < min_aic:\n",
        "        min_aic = arma.aic\n",
        "        min_aic_index = (AR_order, 0, MA_order)\n",
        "      if arma.bic < min_bic:\n",
        "        min_bic = arma.bic\n",
        "        min_bic_index = (AR_order, 0, MA_order)\n",
        "  if verbose:\n",
        "    print(\"Minimizing AIC order: \", min_aic_index)\n",
        "    print(\"Minimizing BIC order: \", min_bic_index )\n",
        "    print(\"matrix of AIC\", aic_matrix)\n",
        "    print(\"Matrix of BIC\", bic_matrix)\n",
        "  return min_aic_index, min_bic_index, aic_matrix, bic_matrix"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 25,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "7VXqpZJFeVKL",
        "outputId": "6341ab5c-3b51-4e31-890b-bea325de43c6"
      },
      "outputs": [
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/statsmodels/tsa/base/tsa_model.py:471: ValueWarning: No frequency information was provided, so inferred frequency D will be used.\n",
            "  self._init_dates(dates, freq)\n",
            "/usr/local/lib/python3.7/dist-packages/statsmodels/tsa/base/tsa_model.py:471: ValueWarning: No frequency information was provided, so inferred frequency D will be used.\n",
            "  self._init_dates(dates, freq)\n",
            "/usr/local/lib/python3.7/dist-packages/statsmodels/tsa/base/tsa_model.py:471: ValueWarning: No frequency information was provided, so inferred frequency D will be used.\n",
            "  self._init_dates(dates, freq)\n",
            "/usr/local/lib/python3.7/dist-packages/statsmodels/tsa/base/tsa_model.py:471: ValueWarning: No frequency information was provided, so inferred frequency D will be used.\n",
            "  self._init_dates(dates, freq)\n",
            "/usr/local/lib/python3.7/dist-packages/statsmodels/tsa/base/tsa_model.py:471: ValueWarning: No frequency information was provided, so inferred frequency D will be used.\n",
            "  self._init_dates(dates, freq)\n",
            "/usr/local/lib/python3.7/dist-packages/statsmodels/tsa/base/tsa_model.py:471: ValueWarning: No frequency information was provided, so inferred frequency D will be used.\n",
            "  self._init_dates(dates, freq)\n",
            "/usr/local/lib/python3.7/dist-packages/statsmodels/tsa/statespace/sarimax.py:978: UserWarning: Non-invertible starting MA parameters found. Using zeros as starting parameters.\n",
            "  warn('Non-invertible starting MA parameters found.'\n",
            "/usr/local/lib/python3.7/dist-packages/statsmodels/tsa/base/tsa_model.py:471: ValueWarning: No frequency information was provided, so inferred frequency D will be used.\n",
            "  self._init_dates(dates, freq)\n",
            "/usr/local/lib/python3.7/dist-packages/statsmodels/tsa/base/tsa_model.py:471: ValueWarning: No frequency information was provided, so inferred frequency D will be used.\n",
            "  self._init_dates(dates, freq)\n",
            "/usr/local/lib/python3.7/dist-packages/statsmodels/tsa/base/tsa_model.py:471: ValueWarning: No frequency information was provided, so inferred frequency D will be used.\n",
            "  self._init_dates(dates, freq)\n",
            "/usr/local/lib/python3.7/dist-packages/statsmodels/tsa/base/tsa_model.py:471: ValueWarning: No frequency information was provided, so inferred frequency D will be used.\n",
            "  self._init_dates(dates, freq)\n",
            "/usr/local/lib/python3.7/dist-packages/statsmodels/tsa/base/tsa_model.py:471: ValueWarning: No frequency information was provided, so inferred frequency D will be used.\n",
            "  self._init_dates(dates, freq)\n",
            "/usr/local/lib/python3.7/dist-packages/statsmodels/tsa/base/tsa_model.py:471: ValueWarning: No frequency information was provided, so inferred frequency D will be used.\n",
            "  self._init_dates(dates, freq)\n",
            "/usr/local/lib/python3.7/dist-packages/statsmodels/tsa/base/tsa_model.py:471: ValueWarning: No frequency information was provided, so inferred frequency D will be used.\n",
            "  self._init_dates(dates, freq)\n",
            "/usr/local/lib/python3.7/dist-packages/statsmodels/tsa/base/tsa_model.py:471: ValueWarning: No frequency information was provided, so inferred frequency D will be used.\n",
            "  self._init_dates(dates, freq)\n",
            "/usr/local/lib/python3.7/dist-packages/statsmodels/tsa/base/tsa_model.py:471: ValueWarning: No frequency information was provided, so inferred frequency D will be used.\n",
            "  self._init_dates(dates, freq)\n",
            "/usr/local/lib/python3.7/dist-packages/statsmodels/tsa/base/tsa_model.py:471: ValueWarning: No frequency information was provided, so inferred frequency D will be used.\n",
            "  self._init_dates(dates, freq)\n",
            "/usr/local/lib/python3.7/dist-packages/statsmodels/tsa/base/tsa_model.py:471: ValueWarning: No frequency information was provided, so inferred frequency D will be used.\n",
            "  self._init_dates(dates, freq)\n",
            "/usr/local/lib/python3.7/dist-packages/statsmodels/tsa/base/tsa_model.py:471: ValueWarning: No frequency information was provided, so inferred frequency D will be used.\n",
            "  self._init_dates(dates, freq)\n",
            "/usr/local/lib/python3.7/dist-packages/statsmodels/tsa/base/tsa_model.py:471: ValueWarning: No frequency information was provided, so inferred frequency D will be used.\n",
            "  self._init_dates(dates, freq)\n",
            "/usr/local/lib/python3.7/dist-packages/statsmodels/tsa/base/tsa_model.py:471: ValueWarning: No frequency information was provided, so inferred frequency D will be used.\n",
            "  self._init_dates(dates, freq)\n",
            "/usr/local/lib/python3.7/dist-packages/statsmodels/tsa/base/tsa_model.py:471: ValueWarning: No frequency information was provided, so inferred frequency D will be used.\n",
            "  self._init_dates(dates, freq)\n",
            "/usr/local/lib/python3.7/dist-packages/statsmodels/tsa/base/tsa_model.py:471: ValueWarning: No frequency information was provided, so inferred frequency D will be used.\n",
            "  self._init_dates(dates, freq)\n",
            "/usr/local/lib/python3.7/dist-packages/statsmodels/tsa/base/tsa_model.py:471: ValueWarning: No frequency information was provided, so inferred frequency D will be used.\n",
            "  self._init_dates(dates, freq)\n",
            "/usr/local/lib/python3.7/dist-packages/statsmodels/tsa/base/tsa_model.py:471: ValueWarning: No frequency information was provided, so inferred frequency D will be used.\n",
            "  self._init_dates(dates, freq)\n",
            "/usr/local/lib/python3.7/dist-packages/statsmodels/base/model.py:606: ConvergenceWarning: Maximum Likelihood optimization failed to converge. Check mle_retvals\n",
            "  ConvergenceWarning)\n",
            "/usr/local/lib/python3.7/dist-packages/statsmodels/tsa/base/tsa_model.py:471: ValueWarning: No frequency information was provided, so inferred frequency D will be used.\n",
            "  self._init_dates(dates, freq)\n",
            "/usr/local/lib/python3.7/dist-packages/statsmodels/tsa/base/tsa_model.py:471: ValueWarning: No frequency information was provided, so inferred frequency D will be used.\n",
            "  self._init_dates(dates, freq)\n",
            "/usr/local/lib/python3.7/dist-packages/statsmodels/tsa/base/tsa_model.py:471: ValueWarning: No frequency information was provided, so inferred frequency D will be used.\n",
            "  self._init_dates(dates, freq)\n",
            "/usr/local/lib/python3.7/dist-packages/statsmodels/tsa/base/tsa_model.py:471: ValueWarning: No frequency information was provided, so inferred frequency D will be used.\n",
            "  self._init_dates(dates, freq)\n",
            "/usr/local/lib/python3.7/dist-packages/statsmodels/tsa/base/tsa_model.py:471: ValueWarning: No frequency information was provided, so inferred frequency D will be used.\n",
            "  self._init_dates(dates, freq)\n",
            "/usr/local/lib/python3.7/dist-packages/statsmodels/tsa/base/tsa_model.py:471: ValueWarning: No frequency information was provided, so inferred frequency D will be used.\n",
            "  self._init_dates(dates, freq)\n",
            "/usr/local/lib/python3.7/dist-packages/statsmodels/tsa/base/tsa_model.py:471: ValueWarning: No frequency information was provided, so inferred frequency D will be used.\n",
            "  self._init_dates(dates, freq)\n",
            "/usr/local/lib/python3.7/dist-packages/statsmodels/tsa/base/tsa_model.py:471: ValueWarning: No frequency information was provided, so inferred frequency D will be used.\n",
            "  self._init_dates(dates, freq)\n",
            "/usr/local/lib/python3.7/dist-packages/statsmodels/tsa/base/tsa_model.py:471: ValueWarning: No frequency information was provided, so inferred frequency D will be used.\n",
            "  self._init_dates(dates, freq)\n",
            "/usr/local/lib/python3.7/dist-packages/statsmodels/tsa/base/tsa_model.py:471: ValueWarning: No frequency information was provided, so inferred frequency D will be used.\n",
            "  self._init_dates(dates, freq)\n",
            "/usr/local/lib/python3.7/dist-packages/statsmodels/tsa/base/tsa_model.py:471: ValueWarning: No frequency information was provided, so inferred frequency D will be used.\n",
            "  self._init_dates(dates, freq)\n",
            "/usr/local/lib/python3.7/dist-packages/statsmodels/tsa/base/tsa_model.py:471: ValueWarning: No frequency information was provided, so inferred frequency D will be used.\n",
            "  self._init_dates(dates, freq)\n",
            "/usr/local/lib/python3.7/dist-packages/statsmodels/base/model.py:606: ConvergenceWarning: Maximum Likelihood optimization failed to converge. Check mle_retvals\n",
            "  ConvergenceWarning)\n",
            "/usr/local/lib/python3.7/dist-packages/statsmodels/tsa/base/tsa_model.py:471: ValueWarning: No frequency information was provided, so inferred frequency D will be used.\n",
            "  self._init_dates(dates, freq)\n",
            "/usr/local/lib/python3.7/dist-packages/statsmodels/tsa/base/tsa_model.py:471: ValueWarning: No frequency information was provided, so inferred frequency D will be used.\n",
            "  self._init_dates(dates, freq)\n",
            "/usr/local/lib/python3.7/dist-packages/statsmodels/tsa/base/tsa_model.py:471: ValueWarning: No frequency information was provided, so inferred frequency D will be used.\n",
            "  self._init_dates(dates, freq)\n",
            "/usr/local/lib/python3.7/dist-packages/statsmodels/tsa/base/tsa_model.py:471: ValueWarning: No frequency information was provided, so inferred frequency D will be used.\n",
            "  self._init_dates(dates, freq)\n",
            "/usr/local/lib/python3.7/dist-packages/statsmodels/tsa/base/tsa_model.py:471: ValueWarning: No frequency information was provided, so inferred frequency D will be used.\n",
            "  self._init_dates(dates, freq)\n",
            "/usr/local/lib/python3.7/dist-packages/statsmodels/tsa/base/tsa_model.py:471: ValueWarning: No frequency information was provided, so inferred frequency D will be used.\n",
            "  self._init_dates(dates, freq)\n",
            "/usr/local/lib/python3.7/dist-packages/statsmodels/tsa/base/tsa_model.py:471: ValueWarning: No frequency information was provided, so inferred frequency D will be used.\n",
            "  self._init_dates(dates, freq)\n",
            "/usr/local/lib/python3.7/dist-packages/statsmodels/tsa/base/tsa_model.py:471: ValueWarning: No frequency information was provided, so inferred frequency D will be used.\n",
            "  self._init_dates(dates, freq)\n",
            "/usr/local/lib/python3.7/dist-packages/statsmodels/tsa/base/tsa_model.py:471: ValueWarning: No frequency information was provided, so inferred frequency D will be used.\n",
            "  self._init_dates(dates, freq)\n",
            "/usr/local/lib/python3.7/dist-packages/statsmodels/tsa/statespace/sarimax.py:966: UserWarning: Non-stationary starting autoregressive parameters found. Using zeros as starting parameters.\n",
            "  warn('Non-stationary starting autoregressive parameters'\n",
            "/usr/local/lib/python3.7/dist-packages/statsmodels/tsa/base/tsa_model.py:471: ValueWarning: No frequency information was provided, so inferred frequency D will be used.\n",
            "  self._init_dates(dates, freq)\n",
            "/usr/local/lib/python3.7/dist-packages/statsmodels/tsa/base/tsa_model.py:471: ValueWarning: No frequency information was provided, so inferred frequency D will be used.\n",
            "  self._init_dates(dates, freq)\n",
            "/usr/local/lib/python3.7/dist-packages/statsmodels/tsa/base/tsa_model.py:471: ValueWarning: No frequency information was provided, so inferred frequency D will be used.\n",
            "  self._init_dates(dates, freq)\n",
            "/usr/local/lib/python3.7/dist-packages/statsmodels/base/model.py:606: ConvergenceWarning: Maximum Likelihood optimization failed to converge. Check mle_retvals\n",
            "  ConvergenceWarning)\n",
            "/usr/local/lib/python3.7/dist-packages/statsmodels/tsa/base/tsa_model.py:471: ValueWarning: No frequency information was provided, so inferred frequency D will be used.\n",
            "  self._init_dates(dates, freq)\n",
            "/usr/local/lib/python3.7/dist-packages/statsmodels/tsa/base/tsa_model.py:471: ValueWarning: No frequency information was provided, so inferred frequency D will be used.\n",
            "  self._init_dates(dates, freq)\n",
            "/usr/local/lib/python3.7/dist-packages/statsmodels/tsa/base/tsa_model.py:471: ValueWarning: No frequency information was provided, so inferred frequency D will be used.\n",
            "  self._init_dates(dates, freq)\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Minimizing AIC order:  (3, 0, 3)\n",
            "Minimizing BIC order:  (3, 0, 3)\n",
            "matrix of AIC [[ 2086.19573446  1200.47398402   400.58363291    99.47049583]\n",
            " [ -906.1707104   -924.25498269  -924.83575415  -922.93914841]\n",
            " [ -920.8804206   -924.20595393  -922.75457547  -976.81167508]\n",
            " [ -923.54108234  -923.93826044 -1007.53084605 -1015.08967408]]\n",
            "Matrix of BIC [[2095.40633031 1214.28987778  419.00482459  122.49698544]\n",
            " [-892.35481663 -905.833791   -901.80926455 -895.30736089]\n",
            " [-902.45922891 -901.17946432 -895.12278795 -944.57458964]\n",
            " [-900.51459273 -896.30647291 -975.2937606  -978.24729071]]\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/statsmodels/base/model.py:606: ConvergenceWarning: Maximum Likelihood optimization failed to converge. Check mle_retvals\n",
            "  ConvergenceWarning)\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "                               SARIMAX Results                                \n",
            "==============================================================================\n",
            "Dep. Variable:              JHU_cases   No. Observations:                  739\n",
            "Model:                 ARIMA(3, 0, 3)   Log Likelihood                 484.519\n",
            "Date:                Sun, 28 Aug 2022   AIC                           -953.038\n",
            "Time:                        00:32:03   BIC                           -916.196\n",
            "Sample:                    01-01-2020   HQIC                          -938.832\n",
            "                         - 01-08-2022                                         \n",
            "Covariance Type:                  opg                                         \n",
            "==============================================================================\n",
            "                 coef    std err          z      P>|z|      [0.025      0.975]\n",
            "------------------------------------------------------------------------------\n",
            "const          7.5413      4.684      1.610      0.107      -1.639      16.722\n",
            "ar.L1          0.0437      0.026      1.660      0.097      -0.008       0.095\n",
            "ar.L2          0.0949      0.026      3.669      0.000       0.044       0.146\n",
            "ar.L3          0.8588      0.028     30.461      0.000       0.804       0.914\n",
            "ma.L1          1.1872      0.035     34.096      0.000       1.119       1.255\n",
            "ma.L2          0.9709      0.044     22.007      0.000       0.884       1.057\n",
            "ma.L3          0.0559      0.032      1.723      0.085      -0.008       0.120\n",
            "sigma2         0.0155      0.000     34.491      0.000       0.015       0.016\n",
            "===================================================================================\n",
            "Ljung-Box (L1) (Q):                   0.00   Jarque-Bera (JB):               972.02\n",
            "Prob(Q):                              0.97   Prob(JB):                         0.00\n",
            "Heteroskedasticity (H):               1.94   Skew:                             0.63\n",
            "Prob(H) (two-sided):                  0.00   Kurtosis:                         8.48\n",
            "===================================================================================\n",
            "\n",
            "Warnings:\n",
            "[1] Covariance matrix calculated using the outer product of gradients (complex-step).\n"
          ]
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 864x864 with 2 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          },
          "output_type": "display_data"
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Automatic selection finds model with AR 3, MA 3\n",
            "MSE with selected model: 0.015533991793349505\n"
          ]
        }
      ],
      "source": [
        "min_aic_index, min_bic_index, _, _ = grid_search_ARIMA(cubic_residuals, range(4), range(4), verbose=True)\n",
        "if min_aic_index == min_bic_index:\n",
        "  arma = ARIMA(log_cases, order=min_bic_index).fit()\n",
        "  print(arma.summary())\n",
        "  arma_predictions = arma.predict()\n",
        "  arma_residuals = log_cases - arma_predictions\n",
        "  arma_residuals = arma_residuals[1:] # Fitting AR 1 model means removing one observation\n",
        "  \n",
        "  fig, axs = plt.subplots(2,1, figsize=(12,12))\n",
        "  axs[0].plot(log_cases[1:], label='AR time series')\n",
        "  axs[0].plot(arma_predictions[1:], 'r', label='fitted line')\n",
        "  axs[0].set_title(\"True Log Cases vs Fitted ARIMA Line\")\n",
        "  axs[0].legend()\n",
        "  axs[1].plot(arma_residuals, 'o')\n",
        "  axs[1].set_title(\"ARIMA Residuals: True Log Cases - Fitted ARIMA Line\")\n",
        "  plt.show()\n",
        "  print(\"Automatic selection finds model with AR {0}, MA {2}\".format(*min_aic_index))\n",
        "  print(\"MSE with selected model:\", np.mean(arma_residuals**2))\n",
        "else:\n",
        "  print(\"AIC, BIC do not agree.\")"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "EdHwj3K5ivvB"
      },
      "source": [
        "AIC and BIC agree with each other on the order of AR and MA which is 3."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "LlcI9DvWdoJT"
      },
      "source": [
        "### Aside: What if AIC/BIC don't agree? \n",
        "\n",
        "One solution is to use a more complicated tool to assess model fit, such as cross validation with a rolling forecast window. "
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "piMAqUNKjm3J"
      },
      "source": [
        "## 8. One Day Forecasting\n",
        "\n",
        "Now that we found a model, we can form forecasts. \n",
        "\n",
        "With estimates of each $\\theta$ and $\\phi$ term, we can predict one step ahead. \n",
        "\n",
        "$$\\hat{X}_{t + 1} = \\hat{\\phi}_1 X_{t} + \\hat{\\phi}_2 X_{t-1} \\dots + \\hat{\\phi}_p X_{t - p + 1} + \\hat{\\theta}_1 \\varepsilon_{t} + \\hat{\\theta}_2 \\varepsilon_{t-1} \\dots + \\hat{\\theta}_q \\varepsilon_{t-q + 1} $$\n",
        "\n",
        "To form long-range forecasts, we can just plug in our estimate $\\hat{X}_{t + 1}$ in place of $X_{t + 1}$ at every future value, and so on for other future estimates.\n",
        "\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Z01hwPrEkDpU"
      },
      "source": [
        "We set 20% of the series as the test set. In practice, we would likely use rolling windows to assess the performance more credibly, since 20% of the series can actually be much further than the model predicts. Basically, we predict on a window and train on everything in the past.\n",
        "\n",
        "An illustration of rolling windows: for $t = 1, \\dots, 100$, train a model on steps $t = 1, \\dots, 20$, then predict on $t = 21, \\dots, 40$. Then train the model on $t = 1, \\dots, 40$ and predict on $t = 41, \\dots, 60$, and so on."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 26,
      "metadata": {
        "id": "Tjzeh4qadSR2"
      },
      "outputs": [],
      "source": [
        "train_test_split = int(len(log_cases) * 0.8)\n",
        "train_price, test_price = cubic_residuals[:train_test_split], cubic_residuals[train_test_split:]\n",
        "train_date, test_date = california.index.values[:train_test_split], california.index.values[train_test_split:]\n",
        "assert(len(train_date) + len(test_date) == len(california.index.values))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 27,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "jGY36En3kpp4",
        "outputId": "72d49f67-fd6e-4341-be92-9ea94079029c"
      },
      "outputs": [
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/statsmodels/tsa/base/tsa_model.py:471: ValueWarning: No frequency information was provided, so inferred frequency D will be used.\n",
            "  self._init_dates(dates, freq)\n",
            "/usr/local/lib/python3.7/dist-packages/statsmodels/tsa/base/tsa_model.py:471: ValueWarning: No frequency information was provided, so inferred frequency D will be used.\n",
            "  self._init_dates(dates, freq)\n",
            "/usr/local/lib/python3.7/dist-packages/statsmodels/tsa/base/tsa_model.py:471: ValueWarning: No frequency information was provided, so inferred frequency D will be used.\n",
            "  self._init_dates(dates, freq)\n",
            "/usr/local/lib/python3.7/dist-packages/statsmodels/tsa/statespace/sarimax.py:966: UserWarning: Non-stationary starting autoregressive parameters found. Using zeros as starting parameters.\n",
            "  warn('Non-stationary starting autoregressive parameters'\n",
            "/usr/local/lib/python3.7/dist-packages/statsmodels/tsa/statespace/sarimax.py:978: UserWarning: Non-invertible starting MA parameters found. Using zeros as starting parameters.\n",
            "  warn('Non-invertible starting MA parameters found.'\n",
            "/usr/local/lib/python3.7/dist-packages/statsmodels/base/model.py:606: ConvergenceWarning: Maximum Likelihood optimization failed to converge. Check mle_retvals\n",
            "  ConvergenceWarning)\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "                               SARIMAX Results                                \n",
            "==============================================================================\n",
            "Dep. Variable:              JHU_cases   No. Observations:                  591\n",
            "Model:                 ARIMA(3, 0, 3)   Log Likelihood                 515.526\n",
            "Date:                Sun, 28 Aug 2022   AIC                          -1015.052\n",
            "Time:                        00:32:06   BIC                           -979.998\n",
            "Sample:                    01-01-2020   HQIC                         -1001.397\n",
            "                         - 08-13-2021                                         \n",
            "Covariance Type:                  opg                                         \n",
            "==============================================================================\n",
            "                 coef    std err          z      P>|z|      [0.025      0.975]\n",
            "------------------------------------------------------------------------------\n",
            "const          0.4289      0.880      0.487      0.626      -1.296       2.154\n",
            "ar.L1         -0.2198      0.148     -1.487      0.137      -0.510       0.070\n",
            "ar.L2          0.7551      0.039     19.201      0.000       0.678       0.832\n",
            "ar.L3          0.4449      0.126      3.533      0.000       0.198       0.692\n",
            "ma.L1          1.5788      0.153     10.344      0.000       1.280       1.878\n",
            "ma.L2          0.9190      0.203      4.521      0.000       0.521       1.317\n",
            "ma.L3          0.0674      0.090      0.752      0.452      -0.108       0.243\n",
            "sigma2         0.0101      0.000     32.726      0.000       0.010       0.011\n",
            "===================================================================================\n",
            "Ljung-Box (L1) (Q):                   0.03   Jarque-Bera (JB):              1620.88\n",
            "Prob(Q):                              0.87   Prob(JB):                         0.00\n",
            "Heteroskedasticity (H):               0.99   Skew:                             1.06\n",
            "Prob(H) (two-sided):                  0.97   Kurtosis:                        10.83\n",
            "===================================================================================\n",
            "\n",
            "Warnings:\n",
            "[1] Covariance matrix calculated using the outer product of gradients (complex-step).\n",
            "Test set mean squared error:  1.2998962464045807\n"
          ]
        },
        {
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAA2kAAAEvCAYAAADfFon+AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nOzdeVhV1frA8e8+h3keBBVBQHMAGRVx1pxzyjQrhxyy7s1bWXm7lt2y4d5u3bKblQ1av0wzm7QyKyvTnLUUnBUFBxxAEZkEBM60fn8cOEmiKSBH8P08j4/sffZe+91HhPPutda7NKUUQgghhBBCCCGuDzp7ByCEEEIIIYQQ4neSpAkhhBBCCCHEdUSSNCGEEEIIIYS4jkiSJoQQQgghhBDXEUnShBBCCCGEEOI6IkmaEEIIIYQQQlxHHOxx0UaNGqmwsDB7XFoIIYQQQggh7C45OfmsUiqgqtfskqSFhYWRlJRkj0sLIYQQQgghhN1pmnbsUq/JcEchhBBCCCGEuI7UOEnTNM1F07Stmqbt0jRtn6Zpz9dGYEIIIYQQQghxI6qN4Y5lQB+lVJGmaY7ARk3TflBK/VoLbQshhBBCCCHEDaXGSZpSSgFF5ZuO5X9UTdsVQgghhKhPjEYjJ0+epLS01N6hCCGuIy4uLgQHB+Po6HjF59RK4RBN0/RAMnAT8LZS6rfaaFcIIYQQor44efIknp6ehIWFoWmavcMRQlwHlFLk5ORw8uRJwsPDr/i8WikcopQyK6XigGAgUdO0qD8eo2naXzVNS9I0LSk7O7s2LiuEEEIIcd0oLS3F399fEjQhhI2mafj7+191D3utVndUSuUDa4BbqnjtPaVUglIqISCgyuUAhBBCCCHqNUnQhBB/VJ2fC7VR3TFA0zSf8q9dgf7AgZq2K4QQQgghro6Hh4e9Q2iQ0tPT+eSTT676vEmTJrF06dKL9i9YsIDMzMwax7V27Vo2b9581cfNnTuXjz76qMbXr6n8/Hzeeecd23ZmZiajRo2yY0RVGzx4MPn5+XV6zdroSWsKrNE0bTewDfhZKfVdLbQrhBBCCCGEXZlMpmonaZdi7yRtypQpTJgwocbXvxSTyXRFx/0xSQsKCqoyqbUXpRQWi4UVK1bg4+NTp9eucZKmlNqtlIpXSsUopaKUUv+qjcCEEEIIIUT1KKWYPn06UVFRREdH8/nnnwNgsVh44IEHaNu2Lf3792fw4MFVfig+dOgQ/fr1IzY2lvbt23P48GGKioro27cv7du3Jzo6mm+++QaA4uJihgwZQmxsLFFRUbZrJScn06tXLzp06MDAgQM5deoUAG+++SaRkZHExMQwevToP72Xl19+mejoaGJjY5kxYwYAO3fupHPnzsTExDBixAjy8vIAuPnmm3niiSdITEykdevWbNiwAYDOnTuzb98+W5s333wzSUlJFBcXM3nyZBITE4mPj7fd04IFC7j11lvp06cPffv2ZcaMGWzYsIG4uDhmz56N2Wxm+vTpdOzYkZiYGObNm2d73x966CHatGlDv379OHPmzEX3s3TpUpKSkhg3bhxxcXGUlJSwevVq4uPjiY6OZvLkyZSVlV103h/ft/T0dObOncvs2bOJi4tjw4YNfPvtt3Tq1In4+Hj69etHVlZWlcc999xzvPrqq9V6L9PT0+nRowft27enffv2tuRv7dq19OjRg1tvvZXIyEieeeYZXn/9dVv8Tz31FG+88Uale5oxYwaHDx8mLi6O6dOnk56eTlRUlO3f4LbbbqN///6EhYXx1ltv8dprrxEfH0/nzp3Jzc0F4PDhw9xyyy106NCBHj16cODAxQP61q1bR1xcHHFxccTHx1NYWAjArFmzbP+Gzz77rO3+2rRpw4QJE4iKiuLEiROEhYVx9uxZAD7++GMSExOJi4vj/vvvx2w2YzabmTRpku3/2+zZsy/+Rr5aSqk6/9OhQwclhBBCCNGQ7N+/394hKHd3d6WUUkuXLlX9+vVTJpNJnT59WoWEhKjMzEy1ZMkSNWjQIGU2m9WpU6eUj4+PWrJkyUXtJCYmqq+++koppVRJSYkqLi5WRqNRFRQUKKWUys7OVi1btlQWi0UtXbpU3XfffbZz8/PzlcFgUF26dFFnzpxRSin12WefqXvuuUcppVTTpk1VaWmpUkqpvLy8y97PihUrVJcuXVRxcbFSSqmcnByllFLR0dFq7dq1SimlZs6cqR555BGllFK9evVSf//735VSSn3//feqb9++SimlXnvtNfXMM88opZTKzMxUrVu3Vkop9eSTT6pFixbZYmnVqpUqKipSH374oWrWrJntemvWrFFDhgyxxTVv3jz173//WymlVGlpqerQoYM6cuSI+vLLL23ve0ZGhvL29q7y/e3Vq5fatm2b7f0NDg5WBw8eVEopNX78eDV79uyLzqnqfXv22WfVrFmzlFJKWSwWlZOTo8xmszKbzWrevHlq2rRpymw2q2eeeUa9/PLLymQyKZPJpGbOnKlefvllZTQaVVRUlFq9erUyGo3qqaeeUlOnTlVGo1H17NlTPfroo8pgMKhvvvlG9enTRxkMBpWfn6/OnTunDAaD2rdvn2rfvr0yGo1q1apVys3NTaWmpiqj0ajS0tJUXFycMhqNymAwqBYtWqisrCxbDCaTSR06dEi1a9fOFvPhw4dt2x988IFq2bKlKigoUFlZWcrLy0u98847ymKxqEcffdT2HvXp00elpqYqpZT69ddfVe/evS9674YOHao2btyolFKqsLBQGY1G9dNPP6m//OUvymKxKLPZrIYMGaLWrVunjh49qjRNU1u2bLGdHxoaqrKzs9X+/fvV0KFDlcFgUEop9be//U0tXLhQJSUlqX79+tmOr+r7uqqfD0CSukS+VCsl+IUQQgghxO+e/3Yf+zPP1WqbkUFePDus3RUdu3HjRsaMGYNer6dx48b06tWLbdu2sXHjRu644w50Oh1NmjShd+/eF51bWFhIRkYGI0aMAKxrPIF1Hbh//vOfrF+/Hp1OR0ZGBllZWURHR/PYY4/xxBNPMHToUHr06MHevXvZu3cv/fv3B8BsNtO0aVMAYmJiGDduHLfddhu33XbbZe9j1apV3HPPPbi5uQHg5+dHQUEB+fn59OrVC4CJEydyxx132M4ZOXIkAB06dCA9PR2AO++8kwEDBvD888/zxRdf2OY9rVy5kuXLl9t6lUpLSzl+/DgA/fv3x8/Pr8q4Vq5cye7du229kAUFBaSlpbF+/Xrb+x4UFESfPn0ue38ABw8eJDw8nNatW9vu5+233+aRRx7B+jne2qkSHR3N2LFjGTZsGMOHD8dgMGAymTCZTLbKgUeOHGHGjBmcPn0ag8FAWFgYBoPB1ttjNBoBa4+q2WwmJyeH/Px8unbtislkYuzYsYwbNw6TyYRSimHDhmE2m4mNjSU9PR2z2UxpaSnTpk1j9+7d6PV60tLSMJlMmM1mEhISCAkJwWQyERwcjJ+fH0lJSZw5c4aYmBi8vLxsMQC26xgMBsD6PVaxbTab6dmzJ05OTjg5OeHl5cWAAQMoKyujbdu27Nmzh5ycHDZv3lxpHltZWRkGg8FWrEPTNLp06cLf//53xo0bx8iRIwkODmblypWsXLmS+Ph4AIqKikhLS6N58+aEhobSuXPni/6tVq9eTXJyMh07dgSgpKSEwMBAhg0bxpEjR5g6dSpDhgxhwIABf/rv/mckSRNCCCGEEH9q8eLFZGdnk5ycjKOjI2FhYZSWltK6dWu2b9/OihUrePrpp+nbty8jRoygXbt2bNmy5aJ2vv/+e9avX8+3337Lf/7zH/bs2YODQ+19JHV2dgZAr9fb5kY1a9YMf39/du/ezeeff87cuXMBa/Lz5Zdf0qZNm0pt/Pbbb7i7u1/yGkop5syZw8CBAyvtX7FixRXHabFYbMmSUuqiBOWPQx6/+uorNm7cyPfff89///tfkpOTba/pdNYZTI899hgPP/www4YNY926dbzwwgvodDo0TUPTNNtxF27/cX9Fe5qm4erqik6nw9HREbPZjE6n46233qJx48YkJSVhsVjw8vKyHe/u7m5rC2Dy5Ml8/PHHZGVlMWnSpEqv/fF6VV3f2dnZ9pper7fFo9frbe+bj48PW7duvejfp+JPxfsybNgwVqxYQbdu3fjpp59QSvHkk09y//33Vzo3PT39kv/2SikmTpzISy+9dNFru3bt4qeffmLu3Ll88cUXzJ8/v8o2rpQkaUIIIYQQtexKe7yulR49ejBv3jwmTpxIbm4u69evZ9asWZSVlbFw4UImTpxIdnY2a9euZezYsZXO9fT0JDg4mGXLlnHbbbdRVlaG2WymoKCAwMBAHB0dWbNmDceOHQOsFfn8/Py4++678fHx4f/+7/+YMWMG2dnZbNmyhS5dumA0GklNTSUiIoITJ07Qu3dvunfvzmeffUZRURGpqam89dZbF1Uc7N+/P//6178YN24cbm5u5Obm4ufnh6+vLxs2bKBHjx4sWrTI1qt2OXfddRevvPIKBQUFxMTEADBw4EDmzJnDnDlz0DSNHTt22HpW/vieVMxjqjjv3XffpU+fPjg6OpKamkqzZs3o2bOn7X0/c+YMa9asYfTo0baEwmKxoJTC3d2dvLw8DAYDLVq0ID09nbS0NFq2bMmnn35Kz549KyU0FouFjIwM2/u2dOlSiouLL4qroKCAZs2aAda5UxU8PDwqHVfB29sbHx8fNm7cSPfu3fnkk0/o0aPHZd/HimvodDoWLVqE2Wy+5LHDhw/n+eefx2g0VllN0tPTk6Kioste73K8vLwICwvjq6++4vbbb0cpxZ49e2z/vhVJ36FDh4iOjiY6Oppt27Zx4MABBg4cyMyZMxk3bhweHh5kZGTg6Oh42ev17duX4cOHM23aNAIDA8nNzaWwsBB3d3ecnJy4/fbbadOmDXfffXe176mCJGlCCCGEEA3MiBEj2LJlC7GxsWiaxiuvvEKTJk24/fbbWb16NZGRkYSEhNC+fXu8vb0vOn/RokXcf//9PPPMMzg6OrJkyRLGjRvHsGHDiI6OJiEhgbZt2wKwZ88epk+fbutxeffdd3FycmLp0qU8/PDDFBQUYDKZePTRR2ndujV33303BQUFKKV4+OGH8fHx4fjx47i6ul4Uxy233MLOnTtJSEjAycmJwYMH8+KLL7Jw4UKmTJnC+fPnadGiBR9++OGfviejRo3ikUceYebMmbZ9M2fO5NFHHyUmJgaLxUJ4eDjffXdxkfKYmBj0ej2xsbFMmjSJRx55hPT0dNq3b49SioCAAL7++muGDx/O6tWriYiIICQkhE6dOmEymWxD/Cp6sCZMmMDUqVNxdXVl/fr1vP/++4wdOxaTyURCQgJ//etfK13fbDZzzz332N63Bx98EB8fH4YMGcKYMWP49ttvmT17Nk8//TRjxozB19eXm2++2Tbc84/HXeiDDz7goYce4vz584SHh/P+++9f9n28//77GT16NIsXL2bAgAGX7XF0cnKiV69e+Pj4oNfrL3rd39+fLl26EB8fz8CBA5kyZcplr12VBQsWMHXqVF566SWMRiN33nmnLUmr8NZbb9mG6bZr145Bgwbh7OxMSkoKXbp0AayJ7Mcff1xlnBUiIyN54YUXGDBgABaLBUdHR95++21cXV255557sFgsAFX2tF0traIbsC4lJCSopKSkOr+uEEIIIcS1kpKSQkREhL3D+FNFRUV4eHiQk5NDYmIimzZtokmTJnaNafr06YwfP/6iD9fXswt7xip6yi5UkZDdyCwWC506deKTTz6hVatWdotDKWUbBmsvVf180DQtWSmVUNXx0pMmhBBCCHEDGTp0KPn5+RgMBmbOnGn3BA2spdCvdxVznC6cR1bhwnldwiolJYXbbruN4cOH2zVBq68kSRNCCCGEuIGsXbvW3iHUGxVJWUViVkGSsj8XERHBwYMH7R1GvSVJmhBCCCGEEOUulZhJUibqkny3Cfv64gtISbF3FEIIIYS4wVksFoxGI2VlZRiNRiwWCzqdzvZHiLok33HCfiwWmDAB/lBlSAghhBCiLlQU/SgrK7MtNVCRlN3oRT+EfclwR2E/2dlQVganTtk7EiGEEELcQCqGNJpMJpRSaJp22dLrQtQ16UkT9nP8uPVvSdKEEEKIWvHmm28SERHBuHHjWL58Of/9738BWLZsGfv377cdt2DBAjIzM6+q7fT0dKKioi67PykpiYcffrgGd3BtKaUwmUy2IY2A9JqJ65L0pAm7UEqhnThh3ZAkTQghhKgV77zzDqtWrSI4OBiAW2+9FbAmaUOHDiUyMhKwJmlRUVEEBQXV6vUTEhJISKhy2Se7qug5uzAxE+J6Jt+hos6VGs30fnUtny7dCIAlKwsuqJ4khBBCiKs3ZcoUjhw5wqBBg5g9ezYLFizgoYceYvPmzSxfvpzp06cTFxfHyy+/TFJSEuPGjSMuLo6SkhKSk5Pp1asXHTp0YODAgZwqf4CanJxMbGwssbGxvP32238aw9q1axk6dCgAzz33HJMnT+bmm2+mRYsWvPnmm7bjPv74YxITE4mLi+P++++vVEWxtlksFgwGAwaDQUrni3pDvktFnVu5Pwt96kGMB1MB0JnNcPasnaMSQggh6re5c+cSFBTEmjVrmDZtmm1/165dufXWW5k1axY7d+7kiSeeICEhgcWLF7Nz504cHByYOnUqS5cuJTk5mcmTJ/PUU08BcM899zBnzhx27dpVrZgOHDjATz/9xNatW3n++ecxGo2kpKTw+eefs2nTJnbu3Iler2fx4sW18h5cSCllq9YIoNfrZVijqDdkuKOoc0ff/5ifP5iBTqnfd546BY0b2y8oIYQQojY9+ijs3Fm7bcbFweuv126bwMGDB9m7dy/9+/cHwGw207RpU/Lz88nPz6dnz54AjB8/nh9++OGq2h4yZAjOzs44OzsTGBhIVlYWq1evJjk5mY4dOwJQUlJCYGBgrd5TxdBGpZQUBBH1kiRpok6VnC9j8vvPUfEcy6zXozeb4fRpu8YlhBBC3KiUUrRr144tW7ZU2p+fn1/jtp2dnW1f6/V6WzXFiRMn8tJLL9W4/T+qKAxiMplkfTNRr0mSJurUqdM5tDCUcLp7X5psXG1N0ECKhwghhGhYrkGPV014enpSWFhY5XabNm3Izs5my5YtdOnSBaPRSGpqKu3atcPHx4eNGzfSvXv3WhuS2LdvX4YPH860adMIDAwkNzeXwsJCQkNDa9RuxfBGi8UivWei3pPHC6JOnckuAKCkTz8YMYLVT/0PgLITGfYMSwghhGjQRo8ezaxZs4iPj+fw4cNMmjSJKVOmEBcXh9lsZunSpTzxxBPExsYSFxfH5s2bAfjwww958MEHiYuLQ104TaEGIiMjeeGFFxgwYAAxMTH079/fVqikuiwWC2VlZVgsFuk9Ew2CVlv/4a5GQkKCSkpKqvPrCvv79rutDBvWidzZb+H36IN8tf0kfbu2hbFj8Z7/nr3DE0IIIaotJSWFiIgIe4dxwzGbzRiNRjRNk8IgokpKqUpDb+2hqp8PmqYlK6WqXLNCHjWIOpV91tqT5uXjAUBjLxdSG4Wi27nDnmEJIYQQoh4ym80YDAZZkFo0OJKkiTqVl3cOAAc3VwAaezmTHByB+55dUFJiz9CEEEIIUY9UJGgy/0w0RJKkiTqVl1tk/cLFBYAATxe2BUeiMxlh2zY7RiaEEEKI+kISNNHQSZIm6lRBfnllqfJxwV4uDuwLbWfdt2mTnaISQgghRH0hCZq4EUiSJuqMUopz+eU9aeVJmqZpuAc1JjuwWe0v+imEEEKIBqVikWqp4CgaOvkOF3WmsMyEVlpm3Sgf7gjQMsCDDM8AyJAy/EIIIYSoWsU6aFLFUdwIJEkTdabEYMbJbLRuXFAGtWWgB+muvqiTJ+0UmRBCCCFq6sCBA8TFxdnWYqtNFQmaUuqKE7T09HQ+++yzWo1DiLoiSZqoM5dM0gI8OOXuD5mZYLHYKTohhBBC1MSyZcsYNWoUO3bsoGXLln96vFIKyxX+3jebzZjN5qsa5njs2DFJ0kS9JUmaqDMlRjPOpvIkrdJwR3dOefqjGY2QnW2n6IQQQoj6LT09nbZt2zJp0iRat27NuHHjWLVqFd26daNVq1Zs3boVgOLiYiZPnkxiYiLx8fF88803tvN79OhB+/btad++PZs3bwZg7dq13HzzzYwaNYq2bdsybtw4lFKVrr1ixQpef/113n33XXr37g3Aa6+9RlRUFFFRUbz++uu2a7Rp04YJEyYQFRXFiRMnmDVrFh07diQmJoZnn33W1uZHH31ETEwMsbGxTJgwAb1ez3fffUf37t1JTEzklltuISsrC4D169fTsWNHOnbsSGJiIoWFhTz99NNs2rSJjh078sYbb1zbN1+IWuZg7wDEjaPEeOnhjlme/taNjAxo3NgO0QkhhBD136FDh1iyZAnz58+nY8eOfPLJJ2zcuJHly5fz4osvsmzZMv7zn//Qp08f5s+fT35+PomJifTr14/AwEB+/vlnXFxcSEtLY8yYMSQlJQGwY8cO9u3bR1BQEN26dWPTpk10797ddt3BgwczZcoUPDw8+Mc//kFycjIffvghv/32G0opOnXqRK9evfD19SUtLY2FCxfSuXNnVq5cSVpaGlu3bkUpxa233sr69evx9/fnhRdeYNOmTXh7e5OTkwNAt27d2LBhA5qmMX/+fP73v//xyiuvMHv2bN544w26du1KUVERLi4uvPDCC8yePZtly5bZ5d9CiJqQJE3UmRKDGWeTwbpxQU+al4sjpY2DrBsnT0L79naITgghhKhdN99880X77rzzTh544AHOnz/P4MGDL3p90qRJTJo0ibNnzzJq1KhKr61du/ZPrxkeHk50dDQA7dq1o2/fvmiaRnR0NOnp6QCsXLmS5cuX8+qrrwJQWlrK8ePHCQoK4qGHHmLnzp3o9XpSU1Nt7SYmJhIcHAxAXFwc6enplZK0P9q4cSMjRozA3d0dgJEjR7JhwwZuvfVWQkND6dy5sy2WlStXEh8fD0BRURFpaWns2rWLO+64Az8/P4xGI/7+1oe5GRkZjBs3jtOnT2MwGAgLCwOga9euPP7444wePZrbbrvNFqsQ9ZUkaaLOXGpOGoB7eHPrF1LhUQghhKg25wt+v+p0Otu2TqfDZDIB1rlgX375JW3atKl07nPPPUfjxo3ZtWsXFosFlwseqF7Yrl6vt7VVHRWJW0UsTz75JPfff3+lY+bMmVOpmmOFadOm8fDDDzNs2DDWrVvHCy+8AMD06dMZNGgQP/74I7179+bbb7+tdnxCXA8kSRN15lLDHQH8W4Zg0nToT55EiuoKIYRoCC7X8+Xm5nbZ1xs1anRFPWfVMXDgQObMmcOcOXPQNI0dO3YQHx9PQUEBwcHB6HQ6Fi5ciNlsrvY1evTowaRJk5gxYwZKKb7++msWLVpUZSwzZ85k3LhxeHh4kJGRgaOjI3369GHEiBE8+OCDBAQEkJubi5+fHwUFBTRr1gyAjz/+2NbO4cOHbfPfkpKSOHjwICEhIRQVFVX7HoSwJykcIupMReEQpdOBQ+XnAy2aeHPGw4+yo8fgl19gwwY7RSmEEEI0bDNnzsRoNBITE0O7du2YOXMmAA888AALFy4kNjaWAwcOVOrxulrt27dn0qRJJCYm0qlTJ+677z7bkMYLDRgwgLFjx9KlSxeio6MZNWoUhYWFREZG8vjjjzNgwAASEhJ4/PHHAXj66acZM2YMnTt3tg2BBGvPW3x8PB06dMDR0ZFbbrmF6Oho9Ho9CQkJUjhE1DvaH6vzXHUDmhYCfAQ0BhTwnlLqsv8TEhISVMVEVHHj+GhLOiWPPsZf9/yAdv58pdfWp2ajG9CfOC8Nj5IicHKCffvsE6gQQghRDSkpKURERNg7jAbBZDJhMpmuquS+EJeilKo0ZNceqvr5oGlaslIqoarja+M73wQ8ppSKBDoDD2qaFlkL7YoG5nzFnLQq/pO0DPRgb+OWOO/bA4cOwf79Mj9NCCGEuAEppSRBEze8Gn/3K6VOKaW2l39dCKQAzWrarmh4bNUdL5iIXKGplwv7GrfE0XLB+PdVq+owOiGEEEJcD650gWshGrJafUShaVoYEA/8Vpvtioah1GjGzWJCq6InTafTaNy7m20738UD008/1WV4QgghhLgOmM3mShUdhbgR1VqSpmmaB/Al8KhS6lwVr/9V07QkTdOSsrOza+uyoh4pMZpxtRir7EkDePqRYeDpSXH4TWwMi8e8dn0dRyiEEELUTE3n+t/oLBYLFotFkjTRoFTn50KtJGmapjliTdAWK6W+quoYpdR7SqkEpVRCQEBAbVxW1DMlBjOuFlOVc9IA0OlgyhS47z6Sm7XF+VSGdXFrIYQQoh5wcXEhJydHErUakKGOoqFRSpGTk1Np3cErUeN10jTro44PgBSl1Gs1bU80XOeNZpwvl6QBvPIK7sCJzHdgNajNm9HuvLPOYhRCCCGqKzg4mJMnTyIjhqqvYpFs6UkTtaXioYmDg/2Wh3ZxcSE4OPiqzqmNaLsB44E9mqbtLN/3T6XUilpoWzQgpQYzrmYjuHj86bExQ3rC26DddRfm5O3oX/5vHUQohBBCVJ+joyPh4eH2DqPeMpvNHDp0CE9PT3uHIhoQi8WCwWCgRYsW9g7lqtRGdceNSilNKRWjlIor/yMJmrhIidGMs6XqEvx/NPWWdhyLaA+A/pWXYb3MTxNCCCEasrKyMnuHIMR1QxagEHWmxGjGyXRlSZqmaYSu/Ia3Xv2cTK8A1FNP10GEQgghhLCX8+fP23VImhDXE0nSRJ253DppVQoOJqRvN5a37YHa+huUj1MXQgghRMNTXFwsSZoQ5SRJE3WmxGjG0XxlPWkVopt5k9YoFJ3BAIcPX8PohBBCCGEvSinKysokSROinCRpos6UGMw4Ga+iJw0I83fnZFD5RM+9e69RZEIIIYSwJ6PRiFJKqjoKUU6SNFFnSoxmHK5wTloFnU7DJTYKi6ZJkiaEEEI0UEaj8aJ9ZSYLczdn8syPR8kuMtghKiHsR5I0UWdKjWYcjIarStIAQkMacfE69wAAACAASURBVNKniSRpQgghRANVWlqKXq+vtG93ZhEfJWWxKjWf7/bn2ikyIexDkjRRJ4xmC0azsiZpV7nieoivGwcahWKWJE0IIYRokCwWy0VDHU+ds/aeBXo4svZQvj3CEsJuJEkTdaKk1MCr389Gf5XDHQGCfV1J92mKdjQdyleNF0IIIUTDdqrQgF6D22MCSDtbwuc7znDeYLZ3WELUCUnSRJ3I2n+IUXtXWzeKi6/q3BA/NzK9AtCVlUJ29jWITgghhBDXm1PnDAR6OtG/jS9ujjre2JDBCz8fQ8kDW3EDkCRN1Ildu478vhETc1Xnhvi6keEdaN04dqwWoxJCCCHE9er0OQNNPJ1o4unEj/fH8LeuQaw9XMCGIwVVHv/SquN8tVse5oqGQZI0USfSUo5bv/j+e7j77qs618vVgXz/JtaN48ev7sIGA7z4Ipw7d3XnCSGEEKLG9mYUsOVwTrXOPXXOQFMvJwAcdBpj4gPRaXAwu+SiY8tMFlak5LBgWxZmi/S0ifpPkjRxzZUYzGSmn7JuhITAVa6BomkahIVZN6pK0k6fhtdfB4ul0u70s8W8OvLv8NRT8Oab1YhcCCGEEDXx3x8OMOOr3Vd9nsFk4WyxkSaeTrZ9DnqNxp5OZBSUXXT84ZwSzArOFhvZdqKwRjELcT2QJE1ccyfzzuNRXN6T5edXrTZ8mwVy3tm16uGOCxfCtGmQlFRp94q9p2h1pLwipJPTxecJIYQQ4po6nlNMxtlCDCbLnx8MnCs1sTm9gNOFBhTYetIqBHk5kVlgrfr4a/o5fkjJobDMRFp575qTXmNRUhY5xUaKy6TIiKi/HOwdgGj4cooN+JSWP9Xy9a1WG62aeHLSM4CW6ce4cBWVNQfPUPbVem4BWLUKEhNtr609mM1LWYesG/lSulcIIYSoSyazhfu+mM2E7d+R4/gs/i88d9ExZovicHYRP3z5FWdLFF/lN6fYqIho7AZAc9/KFaGDvJ3ZdLSAvPNGHlt+GAXENHUnzM8Fdycdj/QI5sXVxxn2wV6a+zgz747WeLvKx11R/0hPmrjmcosNeJcWYXF2BlfXarURF+JLhlcApYePVtr/1fYM/DPSATCvWoVSipRT58gpKiN7535a5mZYD8zKqsktCCGEEOIqnSooJfpUGgA+r74MBgOLfzvGfQuTUOVTFBZtSaf/a+t44t//44V//I1Db4zHYdM8du7YQf/WPkQ1ca/UZjMvJ3LPm9icfg4F3BEbwO5TxSzfl0MLf1eGtvPnr12aMiK6EacLDcxef7KO71qI2iGPFsQ1l1NswLukEOXje9Xz0SrEhfjwg3cTuh1cCyYTOFi/dfdlFNAi1/oDWNu0iZ+SjvLxq5/w3Kp5vO3gSImjCyZHRzzPnKmt2xFCNFBmo4l9054mqV0XhjV3I6BP92o/WBLihlaegJ3IO0/T0nOU6R1wLitl1aLvOfzlal7Y8Cllz/ngkrKP9WlnCfZ1Y9QrH7Li+29plr+HLRtWc37jt5QyAe2Wl2wl9zVNI8jb2rP27b4c3Bx1TO3RjCaeTvzfb6foEuYFwKSO1mJj2UVGjuRcXGREiPpAkjRxzeUUldG6tAidf/XmowEEeDqTGtkBpx3fw6+/QvfunC0qI+fEafxLzrGtWSQdM/aT9fES5i57ESdN4VRWyjf3/ZPANT/RRXrShBCXYbEo3n7pYx5++2Vsi4RMnw6vvGLPsISon5o3B09PTn60kqjiAn5q3ZVbU9YT/vhU+uVmsL9xC5qkH8Gy+BMOp7gzMtSVCQOjeKDrX3B1deXcuXOsWLGCiIgIAA4cOMA999zD0KFDade1L0o5s/tUMd3CvKxVH9sHMqZ94EVhNHJ3YN/pq1ubVYjrhQx3FNdcbrGBRoZitGrOR6tQ1rsvJp0OVqwAIDk9l7bZ6QD8GtMdgA5ffoiHoQSn3btg3z4yx95DposXltOSpAkhLm19WjaGVasBONShBykBYZgXLLAu4yGEuDoZGXDgABmn8/AynOdAQBgALXMzyO3Rmw9e/ZSDjcMpfeVV+m9fyWMPD0d36pTtdC8vL0aPHk1sbCwAJpOJVq1a8f777/PQ+FFkzruPvDXziWt0+VL7fm6O5JeYMElJflEPSZImrrmcYgN+ZUXVruxYoW2bEJKbRWL87nsAMl59i88/fRKAvJ59MegciMo4SJmvH7RqBZGRNPd356y7L2SfASU/pIVoCM4bTJSWGmr2f9pshu++s7Xx6dbj9MjYhyUmFq9fVvJa74nos7OtxwghqqUg/QQA7eJb8XnPO1BNmuC3bCl392zFwrjBuKWmMHH795jatMUSFHTJdqKjo1m0aBE7d+7ktddeo2N0W8z7VnJbbFMA1q1bx/r16zEajZXO83d3RAF5503X7B6FuFYkSRPXXG6RAe9aSNLim/uwISwOxz272bQtjaZb1lpfuPdeAjrGku5r/QGvj4+3zX1r7udGjps3utJSKJR1U4RoCCa+vY6cVpEwfny1zj94upB/T34Bhg2DNWs4W1TGhj0ZtM9IQdf7ZgK9XNAG3kK+mxeWZd8AsGp/Fnkv/BfWrKnNWxGiQTufYi0aMqRvLHeu+QztxAnw8yO+uS95AwZjQSOkIAuHQbdcUXu+vr7cddddLP38E3bt2om7u7WoyBtvvMGYMWOIjY1l6tSprFixgvPnz+PnZp3Vk3veeLlmhbguSZImrrncYgOe5wurXX6/QmRTL3YHW8enb/3sBxJOHcQ8egz83/9xV+dwtHaRADh0aG87JzzAnbPuPtYNKR4iRL23L7OAkQteodnJw7B48UWL2F+Jr3dk0GL3r9aNvXtJSs+lR9pWHMtKoX9/AG5NaM7G5rEYf/yJ/63Yz5Kn38R35pOoadOkV16IK+Sbssv6RePGaDqdregXwD8n3UxOTAfrxsCBV9226wVFfRYvXsz8+fMZMGAAv/zyC3/5y1/429/+hr+bIwDpp7OrfxNC2IkkaeKayz9XjEvp+Rr3pLk46jF06IhZ0xHy87cEFOag79YVAD93J1r16mg9MC7Odo6HswOWRgHWDSkeIkS9t2xrOsMObMCslf/6Onjwis9VSnEi9zxrD56h8/E91p3797PzRAH3bP8W1bw53GJ9ot8vojE723bEOTuLx4a0Y97XL1Kmd0TbtQt27Kjt2xKiQYo+Vb5WaePGF70W4udGwMNTICgIevas0XVcXV0ZOHAgr7/+Ort27eLzzz/nwQcfxN/dAXNRHvcO6c6QIUN47bXX2LVrF5ZqPNwRoq5JkiauKYtFoXLzrBs17EkDiIsMIbVRc0btWWXd0aXL7y926gR6vfXvC7g3b2b94vTpGl9fCGE/RrOFtB/W4WEoYW7v8qGOmzdf8fkLNqcz8/5ZLP7nsN/XUNy/n1Pb99D5+B60Bx6w/gzB+lBo/DP32c5V//0vzzw1nzIHR8wffFBr9yREQ6MumBcWc9o63JHAiysvAnDvvdYiI25utXZ9BwcHunfvTmJiIn5ujqDT0W/MFPR6Pa+99hqDBw+mQ4cObNq0qdauKcS1IElauZyiMtKyZM5SbSsoMdI0rzw5atasxu092q8VdOtm3WjRAmJifn9x8GA4eRJatqx0jntkWwDUvn01vr4Qwn7Wp2YTeSAZgOwxE8l39cRyhUlaqdHMD1/8wjvfvISH4TwWnY7fgtth2bsP121brQcNHVrpnND4CHj/fdixA+2JJxg4pj8/tuqKedFiKC2t1XsToj5bn5pN+llrqfuvNqXZ9jcvyAJPzytab1Cv19vWQ6stzg46fHz9iBkykeXLl7Nr1y7eeOMNOnfuTFhYGABLly7lrrvu4r333iM1NbXWYxCiuiRJKzfxw62cGDkW+vSxdygNSkZ+Ca3PHrdutGtX4/ZcHPVEvDMLPvsMdu0CR8ffX9Q0aNLkonNCQgM57NeMsq1JNb6+EMJ+vtqeQc+Te7C0i6Jb10h+C26HccWPF89LKyqCo0cr7fpu9ykGb1yGs07DfPgIR9NO8kObbujycumauhWTiyu0bXvxRe+7zzaE+ubWgazpNgSnwgJYtuxa3aYQ9YrBZOGNVz7l6w+tlVCXbEitfMCletH+wNnZ+ZoMQ/RzcyCn2Nq75+/vz6hRo3j33XdpVv7gWNM0zpw5w/PPP0/v3r1JSEjg73//OyaTVIQU9iVJWrkerQLos/k7a+UuGatca5KP5XFTzgmUs7O156s2NG0Kd90FHh5XdHjLQA/2NW4p80iEqMcKzhtJ3bidTkd3orttOD1aNeLndj1xPp0JFwxbKigxkjv9n9Ze9txc2/7NqVkMTd2Mbshg3EJDaNmiKR4drD3xA9J+hfbtbUMdL0Wn03Ab2J9M70DU4k+uzY0KUc/szSxg4aIZTHt8NPznP5QVFFU+oKqHH1VwuKCoSG3yd3fkTOGl1zu8/fbbWbNmDb/99huzZs0iISGB48eP2+L517/+xUsvvcTGjRspKyu7JjEKURVJ0sr1aNXo942rmIguLm9rei7R+SehTZtKVZ3qUmRTL/Y2uQmXzJNw9qxdYhBC1Mx3ezKZ+OtXKEdHeOghXBz1uI4cQYmDM+dG3IHlnntAKZ77Zi/p3/5s7U2bPx+wFgwp/WUtjYpy0e66y9bmwCl3kenbBBeTAYdOiVcUR2LLANaGtceybh3Ik3Yh2HnwFB6GEuvGb79hLrIOe+Rf/4Iff4SlS6+onWuVpEUEupGaXUKp8fIP4IODgxk7dizz5s1j6QUxHzp0iLlz53LXXXfRrl07xo8fzzfffHNNYhXiQpKklesQ6kuJg7N1Y+tW+wbTQCil2HY0lza5J9BqYahjdfm4OVEYEW3dSE62WxxCiOpbvjWdESnr0MaMsQ1rfnZsInsG3UFZSRm6BQs4HN6OSdPHEpFlHeqY/d//UXo4nawffqFVynaUpsGgQbY2o8Mb0fSl56wbHTteURyJ4X5sCo1FX3hOfp4IARzZ9fuDbcup01By3roRH28tre/ickXt6PV6dDpdrc8Jiw/2wGhR7D1dXK3zP/roI/bt28eCBQsYM2YMx44d48CBAwCUlJRw77338t5777F3716pGilqlX26Nq5Dzg56MgODcM08Cr/9BhMn2jukeu9EbglFOfn4nz0FkZF2jcWnVzfOfeCO++uvo6/GeixCCPs5V2rEadMG3MvOw+232/Y76HUkLl/Eyl0niRrai5bHUmyv7R98B21XLEV3UzhNgD5NWmFseRNOnp6V2tbuvRfc3Su1ezlBPq4ci0mE5cAvv1xUTVaIG032gSMAZHo2IvDUKVxblQ8JrEbFRmdnZ0wmE44XzjevodimHug12H6ykIQQzz8/oQoeHh7079+f/uXrKJrNZgAyMjI4ePAgP/74IwA+Pj506tSJRx55hNjY2Nq5AXHDkp60C7jrrE9vDBs22jmShmHnyXwe+HWJdaNHD7vGkhDdnDe73oX+xx9h7VqZdyhEPZJ8LI8+h7ZidnaBvn0ven1AbDCNN63lsSmvUaa3friLfPV5Nt79kO2Y2NNpOF6w0L2NgwPcfTc4OV1xPCFtQjnSJBw2yu8KcWMrKDHifMZawXlvk5vQncnC1Vhe+bQaSZqLi0utF+xwd9bTJtCN9UcKMJkv7qUzWRSLk7N4asVRTJYr68XTl89fvemmm9i4cSPbtm1jzpw5DBo0iIMHD9p61NasWcPkyZOZO3cuSUlJGAyXnhsnxB9JknYBN6xPRpz27oFTp+wcTf13LHk/D2xZYp0r0quXXWNJDPfjkw5DrcOdVq7E0iiA72+fwu6lP0FBgV1jE0Jc3rajufQ+kmytvnuJD3765iHcO/Ne8voNAm9vaN2abgvf4IEPf+WcszsAWnx8rcQT2dSLfT7BWFIO1Ep7QtRX6WeLaVKYA8Cexi3RGY227eokaW5ubrZeqto0tn0gR3JK+b/ffv9sl5Z9nk1HC/hqdzZvb8pkzaF8MgqqVxgkKCiIkSNH8uqrr7Jp0ybiyivC5uXlcfDgQf79738zfPhwIiIiuP322yko/9whwyPF5UiSdgEHo4G9IRHWjR9+sG8wDcC5nXvQodDdd9+fH3yNebo40ioskGzfxhiXf4cuL5chX80j5o5b4MEH7R2eEOIydqecICwvE323rpc9LjLIiyYffwDr1oFej16n8daERJx7lz8kKv/gVFORQV4c8QtGO5YOUu1N3Kjy8jiWVUCTwhxMbu4c9bOWtG9eUL42ajWStNoc5nihPq186dvKhy93Z2MyK9YdzmfipweZ/u0Rfj12znbcsdxSLEqRln2+RtfTNA2AkSNHsmnTJnbs2MF7773H+PHjcXFxwcvLC4DHHnuMPn368MQTT7B06VKOHTsm67QJG0nSLqCVlXEuKpbTXo2wfPedvcOp18wWheFI+TpFISH2DaZc15b+HPBqiuO+PbZ9Bp0DaunSSqW6hRDXj1KjmbJde60b0dF/fkKjRnDBXBCdTsN5yCDrcMb2VQx3rAZrkhaEZrHA4cO10qYQ9YrFAn5+tHryERoX5UCzZmS7+wLQPK9mSdq1KB4C0LeVL8UGC3tPF7PyYJ5t/9bjhXQsn6t2LK+MZXvOMvHTg5zMr70HMIGBgQwZMoTnnnuOxYsX25K49u3bExQUxDfffMMjjzxC165dGTlypO283bt3k5eXd6lmRQMnSdqFyspoEujDL+EJqJ9/Bhk7XG1Hsovwzz2DRa+3rmt2Heja0p903yAAlE7HsV82M3rsS2hlZbB4sZ2jE0JUZdeJfFqUV2skKqp6jUyZAikpEBBQKzE18XLhbFCYdUOWbBE3oiLrWmgRq76heUke+pBgzvlYlzKqSU+apmm4ubldk7lbCSGe6DX49dg5juWVEtXEOgzaoqwVIBu5O3Asr5QVKdaHtmlnS2o9hj8aP348H3/8Mfv27ePnn3/mpZde4q7yZUKUUowZM4aoqCi6d+/O1KlTmT9/PqmpqX/SqmgoJEm7UFkZQYHerG2ZgL6oSCaF18DukwU0K8zG3CTIbuuj/VHHMD90rVsBoLVoQWjvLpR17MTJgGDU6tV2jk4IUZVt6bm0yT6GcneHsLDqNeLgAC1a1FpMmqbhGlU+NF4+MIkb0bnfhwg2KcpFCw7GMci6NEbz/OonaWCtpGg0Gmsc4kXtOuuJaurOlvRznMgrIzbInWBv69JLrQNcae7rwsajBezPsg51TM8trfUYLkWv1xMZGcmECRMYPXo0YE3S5s2bx4wZM2jTpg2bNm1i5syZLFliLchWUlLCzJkz+eKLL0hJSan1givC/mrl07OmafOBocAZpVQ1H3XamdkMFgsuHm4UduuCcbkjjt9/b52oLq7a7pP5DC48i0NYc3uHYuPiqOfuCf3h89chwvoBa1ynUH4LbMWtm3/FUSkoH4IAYJk0Ca3MgPbpJ/YKWYgb1sHThTT1ceG3o7n8/dxJ61qLuuvnuWJ4i6accfel0YED8rRT3HBKz+ZSsfpZQO5paNOGwKDGlDo44VNahNLr0ao5v8zV1fWazcvqHOrFvC3W4iGhvi5ENjFysqCMVo3caOrlxPaTRTjpNVwcdRzLq7skrSo6nY7u3bvTvXt3wJq0nTp1yjZUMj09nc8++4z58+cD1sqYERERPPHEE/To0YOy8vmyzs7O9rkBUWO19btlAXBLLbVlHxWTv52d6RgdypaQKMzff2/fmOqx3RkFhJ7PQWt+/SRpALSy9qRVrNs2PC6I/cFtcczOghMnbIeZzRYKv/wG9fnnkJlpj0iFuGHtPpHHj6Pup6BZGH/5zwNEHdtXaZ7Z9SCiqRepjZpj2L7T3qEIUef2ppyovGPgQFo29rDNS9Pc3Co99Lwajo6OuLq6XpPetM6hXravQ/1cuC2qEXfEBtDI3YG2gdaev3/2a05EoFud9qRdCU3TCAoKomn5FJKIiAgOHDjAunXreOutt5g4cSKurq64uroC1vL/rVu3ZsCAATz22GMsWLCApKQkW/Imrn+1kqQppdYD9bvywgVJWq/WjdgUGov+4EHIyrJvXPWQ0WwhJSOfRnln4HpL0lq0sM5PGTsWAHdnB862K6/4tnWr7bBlK5LwLspHpywyX02IOmSxKN567wce2fQpxd5+xBachJt7w7PP2ju0SiKDvEgKjsR57+5KQ7+EaIjOlRrp/epaktKtH/X2HrggSQsIgPh4Wjby4JiPdchjdYc6VvD29r4myUSrAFd8XK2DyEJ9nYlr5sG0XsFomsbwdo34bHwEA9r4EebnwrE8a6XH65ler+emm25ixIgRPPPMMyxZsoSEhAQAwsLCmDJlCgEBAfz888889dRTDB8+nIyMDAB++eUXXn31Vb777jsOHz58TZY+EDVTZ5OFNE37K/BXgObX2wd3qJSkxQb78L+WcbAWWL8e7rjDnpHVO6lZhXgW5OFgMl43lR1t9Hp4991Ku8zRMRgcHHH49Ve0d96hzD+A1U5R3A4UO7rg9tXXaNOn2ydeIW4wKafPEfibdT5w25Vfw0032TmiqrUM8GB7aDTapk9h82a4pX4PJhHiclJPF3L0bDE7j+fRIdSXw2kXjDDp3x90OloGurPHP4Tux3ZBeW9OdbnW8PxL0Wka3cO9SDpRhJdL5Y/ADnqN5r7WQZyhfi6UmRSnzxkI8q6fwwXbtm3Lk08+Cfw+VHLv3r2EhoYCkJSUxJw5c2xrtVUMl/z6669xdHTk2LFjeHh44O/vb7d7uNHVWZKmlHoPeA8gISHh+ns0cUGS5qDX4dOtM+cXueC6bh2aJGlX5XB2MeF51ic1lP8wuJ4FN/YhtVFzTMvXEJe2HRegQ6J1mMUvLTsyOC0FvX1DFOKGsT71LN2P7cTcvDn6li3tHc4lOep1nG/fEdNnevTr1qFJkiYasBN55wnOP819vYZS+OUyDHn51hdmzYIRIwBo0ciDr/3LH8zWsGy8k5MTDg4OmM1m9Pra/Q38aM9gig2XX0S6ovLj9oyiepukXahiqGRQUJBt3+OPP87UqVNJS0sjJSWFlJQUcnJybGvVzZw5k9WrVxMYGEjbtm1p1aoVsbGx3H777fa6jRvO9VF273pQkaQ5OQHQLbIJSUERdPplDfX/v2fdOp5TTK8jySgHB7TyCa/Xs1B/Nw42CuXW/ets+yYlf0tpSCj7Grdk2IEN1uFMXl6XaUUIURs2HTjF+ON70N89utpzWurKiO6t2Nm0NZGffsEPQycTumQRCYEu8M9/2js0IWrVydwSpm2yFtEyLfsGj7Ly2TJ//avtd6OvuxOHK5K0goIaX9Pb25u8vDzcajh08o/cnPS4OV0+8Wvp70Ijd0d+TT/H0MiG25Pk6upKTEwMMTExF702depUunfvzoEDB0hJSeHTTz9lz549tiTtzjvvxGg00qpVK2666SZat25N27ZtadKkSV3fRoMlSVqFijU5yqvg9GwVwEdh8fRcOx/S06tf+vkGlJ5znilHt6H16AE+PvYO50+F+rmxrlFzHC3W8di5TYLxzT+Lcegwjmd6WA86evS6K1wgRENTVGai9NdteJQWQb9+9g7nT93RIYQZ/Sfw2oIn6XZLZ5oUlU/NHjgQOnSwb3BC1KLc9JP8bd9aAM4rHV5l1jL1eHhUOm7CvYPgs9p5SOHh4cHZs2drpa2rpWkanUM9WXe4AJNF4aC7vh8YXQsdO3akY8eOtm2LxUJhYaFtu23btuzZs4cVK1bYFtwePnw477zzDgAzZsygadOmtGjRgvDwcMLDw3F3d6/bm6jnaqsE/6fAzUAjTdNOAs8qpT6ojbbrzAXDHQFC/NzY36kPrJ0PX38N06bZMbj6wWJR7Ms8x/nUQ9yUlQ6PT7V3SFekeXlPGoDSNPwOHwQ3N9wsioz75lgPOnJEkjQhrrFlOzLofGS7daNvX/sGcwWcHHQ89MpD7MzZQ/O03SxMuI9bv5yL07THcF/7y3W1ZIAQNZJyAAdlHSJoOXUKD4MjFg9PdH/4Hh/Ut/Z+Tzo7O+Pq6orBYMCpfJRTXeoe7s13+3N58Ms0Zg1rcdEctj/ac6qYZt5O+LlVb+mB651Op8Pb29u2/a9//cv2dU5ODmlpabZez6KiIn7++WdOnz5dqY1//OMfTJs2jZKSEj788EPCw8Np0aIFoaGhuLi4ICqrlSRNKTWmNtqxqz8kaQAtu8RyIDCc1l9+iU6StD/17e5MHvlsJ+O3r7LuGDrUvgFdoaberhxpHAaAFh5uq0ql12mYw8KtBx05YqfohLgxKKX4bOMhXsrcg4qPR2vUyN4hXZEWAR6w3FoBdmhRGXMLSnny2zexPPssun//287RCVE7HI8dBSDLOwD9mTN44Ivm433xgZoGd95pW4u0pvz8/MjMzLRLktajhTf/uDmYV9eeZOXBPEbFBlzyWJNZcf+SVFr6u7BoXO3ce33i7+9fqcCIh4cHycnJFBcXc/ToUdufip659PR0/vOf/9iOr5gz9/zzzzNo0CByc3PZtm0bISEhNG/eHI8/9NjeKGS4Y4UqkrSerQJY2TKRNluWyJykK5BdaH0P+x3aSl5wOL6tW9s5oiuj12ncOaIrpgUeOLRrV+k1/+DGFLp64ilJmhDX1Kp1+/j0qWF4GkqgnlZT9fdwJvL56Xx5dC8jX3wRxo2Dtm3tHZYQNWIyW/DOPI5J07GzSSs6nM3Az8cF7VKfiT7/vNau7ebmhqZpKKVsizjXFU3TGBkTwJe7z7LmUP5lk7STBdbPP9fb2mr25u7uTlRUFFFRUZX2R0REkJKSYkvejhw5wtGjRwkIsL7H27dvZ/Lkybbj/fz8CA0N5cUXXyQmJobMzEwOHz5M8+bNCQoKshU7aWgkSatQRZLWuaU/C8Jj0bZ8Dhs2wJAhdgqufigoMeJmKKHz8d2k33UPvvYO6Co81Lc1zP8AwsMr7e/a0p+j3o1pfTAN6YgX4tooLjOxau7/s3fe4W3V99u+n+RNkQAAIABJREFUj7YlWZb3XnHi2FnOHmRDwiZhhF2gjFIolE0ZLT86oS20QEsLpW+hBcoIBcLMBJIACdk7znKceO+hvc/7x5FlO3bIsiyPc19XLnSW9MjY0nm+n/Uu8z1OArfeiuLuuyMt6bS5pCidG669l/N/tR71I4+h+ejDSEuSkTkjKpqdZDTXUBebTJUxnllHtxMXnQjx3UTSepi2FDur1Rq2tvwnYu5QM69trKHJ4T1uKuPhRicACcaBaRbCgclkoqioiKJuSknOOussPvvsM8rKykL/ysvLQ+mUq1atCo0XUCqVpKWlkZmZyfPPP096ejqlpaXU19eTkZFBcnJyrxv8nkI2aW0c090RwKhVwdRpeN9Vo169WjZpJ6C21cljq19D6/cRf10/HFtw1VVddp1TmMT2hGyGb9wATucZz36RkZHpypf76ig8uB2/Xo/ypZegH6+KKhQCj9w4i1dWLOL+j9+At9+GBQtg61aYOTPS8mRkTpnvDjdS0FKNOyuHekMsUS4HifZmyO2dGYZtXR4jxdyhZl7dWMPaklYuHd19GvbhRimCljBA69F6G71ez9ixYxk7dmy3xy+++GLy8vIoLy/n6NGjlJeXU1ZWFkqLfPfdd/nrX6WeAgqlisycXL5csazX9PcUsklr45jujm1MG5XB1rThjP/yK+Q/ve8nftM6btj2OTz0EAkX9v3ObCdDXqKRF6ddwKLdX8AHH0jpSzIyMj3K8j013FO5F8VZ0/u1QWtjTIaZTU/+gk03bGH0bbezfsS/mbt5hWTYrrkm0vJkZE6JtQfrOb+1BsesKdRbpI7NqXUVYBrfK6+v1WoxGAy43W602t4fijQkXkeWWctXh1qOa9JKg2mOHn/fGwM8EImLi2P69OnHPX7jjTcyZcoUysor+P1HW2gOdu/ub8itp9roJt0RYOawBNZljUG1bStEqBVsf0FTflR6cPfdfX6+0ckiCAKJl5xHmTkF70v/iLQcGZkBh9vnZ/u2QwytK0WYPSvScnqMa84awrMX3onOYWPu5hUEEAjcehs0NUVamozMSePzB9i+6yhmhwXF0KHUG6RCBrXXDTHhT3dsIy4uDk/bYnovIwgCc4aa2VphpcXp6/ac0mAkze7pn2ZgoJGWlsbcuXOZceEiYmffROzcW058UR9ENmltHMekjUqLYcuIKQiiCCtWREBY/0HdUC89SDx+cW1/5NIJmbwx9kLU334NW7ZEWo6MzIDio21VXLR5KQpRhIULIy2nxzBoVUxZNJ+t2VLB/LPn/giFwy6NdJGR6Sfsq7ESW1sBgKEwnwZDh2rzXmymFhUVhVarxev19tprdmRefix+Ef67tbbLMYfHT3mLZNKsbtmk9SX21Unz/PrrnDvZpLVxHJOmUAjEzZ5Ok8GM+PnnERDWP/D6A+hamvDo9KEW9gOFEakmNp+7CLvOgPjMM5GWIyMzYLBbHVh//gS3b/sUcc4cGD060pJ6lPvn5zPirX/Co48S/ejDlMam4nz9zUjLkpE5aUrqbWQ1VwMQPWI4f/7pee0He9GkCYJAQkICLldkuicOTYjikpHxvLO1LtQkpI2dVXb8IhQm67F7/IiinPLYV9gfNGnRWmWElZwesklr4zgmDWBWQTJf5YzD//lS8MurJN3RYHMT72jFExd/4pP7GYIgcPmcQt4ZNQ/xgw+gsbH9YCAQOWE9wQcfyGm8MhGhzuriibue49aV/yHO0ojwyCORltTjCIKA7qyp8PTTnDcqhY8LZ6P9Zi0cM+BVRqavcqTBQXZr8Pc1L4/88cPhoYdg0SK49NJe1WIwGFCr1fh83acchps7pqXiF2Ht4VYANpdbOdrsYkulFZVC4KwcEwERHN5+fl8wgNhXJxlqi9vXL82zbNLa6Ka7Yxuz8xP5Km8SquYm2Lixl4X1D2otkknzx/ePAbSnypUTM/lyyvkovF547z02bj5A+cTpiLm57U1n+hFOj5+fPvU+XHEFPP54l+MOjw+vX/6ikQkf60saGXpwOwG1GqG1Fc4/P9KSwkpugoENk+ajCATgvfciLUdG5qQobbBR4KiHhAQpciYI8Mwz0u9wN63Tw4kgCCQmJuJ0Ok98chiI1avJjdOxo9JGi9PHPR8e4to3ivnvljqGJ0WRYJCaHtnklMc+Q9vcOn8AbJ7+d08jm7Q2jtPdESAxWkv9WbMJCAqQUx67pdbiIt7RipCcFGkpYUGnVjJp4VwOJGTR9MprHLzjATK3rEMoK+sfxv3QoU6baw7UE/X1WgDEd94Buz10zO728ZNH/sNz/+h/7Wpl+g/F1VYmVRbDuPEQHR1pOWFHEATSp09gf3Iu4ttvR1qOjMyJ8fsZ8ul7FDaVw5AhkVYDgNFoRKPRRKw2bWy6kV3Vdhrs0uunmaSF/Zm5MaGUOpvcPKRP4PD4sXn85MZJU26P1/SlLyObtDbcblAqpX/dMGlsHlvSC/F/+lkvC+sfVDQ7iXO0ok1NibSUsDFvRApLRswhbttGLtm7ls1DxhIQBPjyy1N/Mp8Pein07v7iSxg2DNZKpgyvF/O9P+GOje/jUagQrFb43/9C53/yzGv846938MBPFxAIzhmRkelpDpU3UFR9EMWM47dRHmjMyk9kScEshPXrobw80nJkZL4XcfVq7vnv0ww/tBPy8iItB5AWO5KSkiIWTRuTasDhDbCl3ArAY+dk8dEtI7luQjIGTdCkyZG0PkF90EjnJ0rzbVtd/e//i2zS2nC7u42itTFneCJfDZmAcvs2uZ6gG8oabCQ4WtGkJkdaStgYmWbi28nnAmByWrEtupo9SUNwr/zi1J6ouRmSkuCdd8KgsjMbDjey+JHnpI2tW3F5/Tz71FtM/fJDhjRVsqVwMq3RZvjqKygqovZ3zzD7j49TkZDBuqwxiD/7GfX7SsKuU2bw4d+2HY3PA98z62agMa8wmXUFU6WNVasiK0ZG5gTYLO0ZFn0lkgbSoGO9Xo+7rUylFxmZYgBgY5lk0sxRKhKNGlQKoT2S1o1JK651sKfG3mW/TPiot0kmbZhs0gYAJzBpYzPNbCqcIm3Irfi7UFdZjzrgQ0gamOmOIK3gTZw9jm25YwBIuvoy1mUXodq4ARyO0Hler4+ST76ArVu7PIfL6+eVX/5TMmrh/D3y+2HzZl5ZU8K8/eulfcXFbC1rxvLN+tBpDVdez5akofg/+BB27qT5tTdJtTZgvP1WHj/vLnxeH1ULr4KqKumCQ4fkMQQyZ0yT3UN8WdD8D7COjt9HlEZJ/tlTqTfG4lsuf4/I9G2qyuvaNwyGyAk5hrZoWiTmpiUZpbqzgw1SJC8mShU6Zvwek3bru/v50eIDvaBQpo02k5afKHUcb3HJ6Y79lxOYNJVSQdLMKTQY4xCXybU6x2IrC97ED2CTBvDYBQUU/ufv8Pzz5BcNY0P+RJQed2hV3O3z89pPnyZvwTyYMAG++67T9etLGjGs/QoAcdOmsOm0v/EWTJrE5X98kFRbI35BAcXF7K5spajmIIHkFAgEKLjjBnYlD0NptQCQX7ILgOSp40kYXcAv591B/uE9BK7/gfTE99wjdfRqbYXFi+GLU4wiysgA+2os5DVWEFCrITc30nJ6lcsnZPJ19lj8K1b2/+6wMgOaQyVS633f5VfAbbdFWE1ndDodJpMJR4cF0t5Ao1Jg1qlCNWkxuvYSmeOZtEA/7Co4EKi3SSZ+aEIwkuaUI2n9F7e7286OHZkzPInVOePwL10mt+LvgD8g4qkJDngcYIOsj0WlVKCbOR3uvReVUoF/+kxsOgPVT/yW/VfexLorbqPos3epM8RK9WrHpDSt3l/H9CM7pI3iYrDZelyjy+tn2dsrAbho3zc0ZA/jk8KZ+IuL2VVpYVxdCYpJE0EQGJYczYgFZ4euVRD8Mhkxgrd/NJUpTz/KvyYtRPh6LY2Vdfi3bIGKCg5Omg1XX43//AugsrKLBs++A9KXeoTqBmT6NvuqrQxpqiAwJA9UqhNfMICYkhvH7hFT0DY3wgcfULZmI960dHj99UhLk5HpRNkR6Xtd9a//1ye/2xMSEggEAgR6ebEjwSh9Zhk0CtTK9tvoUE2ax8/uajsVLVI6Zo2l/3WAHgjU271Ea5WYo1ToVApa5HTHfozH872RNIDZwxNZkzseVUszbN7cS8L6PjUWFzHWZmljgEfSjmXy8BS+yRpD6s5NZC15l9kf/4cpFXvYdv6VHEjKJfDVV6FzRVFkz4Y95LRUs3X4RIRAAP/6777n2U+PD7dVYiw9SECphGuuwffRR+xKHoqyoYH63fvJqS+DiRND58+/4UIALNpgOoteD1lZRGmUTMyJZU3ueAS/n8UP/AFlnZT+MuzgDjblTwS/H9/zL3R6/fe3VPDZ1XfBv/4FK1f2+PuT6f/sq7GQ31yJsrAg0lJ6HYVCwHDTdexMGYbzltvYfufDqKur4KabukTeZWQiRXmTA1dj8HvdaIysmOOgVqtJTEzEbu/dWq+2VvvmqM4LTFqVAo1SwOb2c/t7B7jq9b0ERJHSpsgM4B7s1Nu8Hf5fKeXujv2aE6Q7AiRF66ifNkuKkMgpjyGONtqJc0jDHfvials4uW1mLsP++Et8555H+doNbL3/ScSEBMx33sb6jJEEvl0XGu9Q2mAnae92AAI/vQcA5bnz8d9wY49GnDYcbqSwqQzh0kvh7bdJKSrEO2w4AAs++zcKUYTZs9svSEuDhx/myOO/lrYLC0EhfTSkm6OoLCjCrtVzybI3O7/Q/Q+wNP8s+Oc/weejrq6ZVf98n/+8v54L96yRzpFHVsh0w4HKZjKbqxEKBp9JA7h+xjCevOpRFA47C4rXsmHIOHxKJSxZEmlpMjIAHKi1YvA4Ceii+nS0OyYmBo1G06v1aaEbf13Xn0uMTkWTo90MrC1p5XBju0nz+eXUx97gg531rD3cSmKwhvDy0QlMyeqbiw3fh2zS2jgJkwYwYfxQdqTm4/t8aS+I6h9UNDmJH6QmTatSkrfwXFTLl5E/rYiJf34SoaaGSXMncnTMFFRuF4G//AWA1fvrKao+gKjRMP7Wq3j14ed5bcIlKN98A559tsc07ThYQ0ZTNcKoUaF9KefNwaI1cN2O5Xgys2DWrM4X/fGPjPn5vaDTwYgRod2CIHDDzGEsHzqFDEuwiHzGDDCbSb3sApYNPwtVawvili18c9vDzLt9EUuevhp1wE9xYg6+Tz/rtVEDMv0Df0DEsf8QKr8Phg+PtJyIkBKj45F7F/LJ/OsBqLn6BrakFeJdKi/+yUSOxtomlj7yDF6vjwabG6PHiWjq2zMMFQoFKSkpuFwuxF76rkk0SqUxMVFdTVqSUc3BhvY6uSW7Gyhtal+Etcsz1MLO3ho7z66uAAh13LxufBLnDI2JpKzTQjZpbZykSZszPIk1ueNRbt4EjY29IKzvU90qDbIWo6Olm/zBjlKJQiEw4a4bWJY/DcXDD7PpvCtJeOherti3FmHcOBQ6Lbf88V6W3PwI5am5PZY+W9niRH/4oBQtGzkytP/cGQW8POs6ADS33hKKlB2rm3fegZ//vNPuO+fkkfnSc+07Xn8dVqwgPSmGfQUTANj95hIKv5OaiBy44XYOfLiMf0+4BFVlBezf3yPvTWZgcKTRTm51qbRRWBhZMRFk6pB4Fi35B7zzDkX33sranHGod+6A+vpIS5MZpBx88g9c8MefUf6nv1FvdWP0OBBMpkjLOiFRUVHEx8f3WhOR46U7gmTg2iJn8XoVu6rt7KxqT8eUB12Hn7e21mHUKLlhQjJXj+3fJTiySWvjJE3auEwzmwsmIwQC8pybINWtTtI81gHdfv90uHh8JqUv/os3J1zMpBX/Y8GGT0iwNEBRUeicSdmx7DKnE9i794xfL7C3mJabbmPB3mCq4ZgxoWNDk6K5/6O/wPPPw733Hv9JFi7sNroxaWYRfPopvPee1I1v0iQEQSB9eA6HknMwvP0mhfVHCPzxjxS8/jLp581hT3Jw+GkPvDeZgcPuylbGVe8noNHA2LGRlhNZtFq4+mpykk1UTgmmIL/xRmQ1yQxa6qsaAIha/A4NNg8xPheKfmDSAOLi4lAqlXi93rC/VptJi+km3TE5Wk0gGNCblx+L0xugyuJhWrb0c5QHXYcXX0BkTUkLFxTGcef0NEal9p3REaeDbNLaWLz4pL4cVUoF5tnTaY2KllvxAwQCOI5WkOyxDrpUxxMhCAJ3zi/k2o0fs2/TXp68KxiNuuSS0DmTcuM4EJeJcPjwGdWlfbC1gnevu5+RH7/Fjzd+gHj55V3MljpKJxk0s/n0XuSii2DRok67zilMYm1mEUMapdQCxcKFAETr1LhygsNP9+07vdeTGZB8d7iJiTUHEMaOPamFscFCwUWzWZszDv9vfgtNTZGWIzPI8AdE/GXlAKRu+w4OHSTW54Lovp3u2IZSqSQlJQWn0xn2tMfvj6SpQ4/n58cCIADnF8YBcrpjuHF4/PhFSDN9f7f2/oJs0tpITYX09JM6dVZhCmuzx+L/fKlcb/OnP/GXBy9ieFnxoOvseLIoFQIFEwv51Yv3SRHbiy8OHZuYHcvB+CwpMnsGaYFLd1RxzqGN1KZk4R83DuGFF058UQ9w47Qcbl7+Kvz1r/DCC5CfHzqWnpVMQ0yinO4o04lNh+oYXX0QYerUSEvpU1w0Jo0/zPkhypZmePfdSMuRGWTsrbKQ1FxLmTkZj0rDjA9fI9rn7DcmDUCv1xMfHx/2bo/pMRr0agW5cV3LO5KM7eZgaEIUefE6itKMZJmlBanuImkubyDUrl/mzGgzwQat8gRn9g9kk3YazMlPZM2QCajqamHHjkjLiSzffguA3mmXI2knwzGz+OKNWnxtHe727oXVq0/Z1Pj8AWzfrCfJ3kzys0+h3LoVMjJ6SPCJEeLi4O67pUHXHRieEs3B2DRE2aTJBKmzutAW70HrccGUKZGW06fIjNOjGj+OBnMirF0baTkyg4xt5c2kW+qoKxjD/8bMZ9b6paTVVUA/SXdsIz4+Ho1Gg9sdPtNj0qn4/EejmZ7b9WeTHN2WCqlEo1LwzII8fnNBTmiGWneRtAc/LuGq1/ceNwLY5Ah/CudAwe6RZuYZNbJJG7QkmXRUTZsjbXz2WUS1RBpvegczIEfSTov8mePxKpTseukNvPPPxTPtLCgtPenrd1S0MmPvt9JctAsvDKPSU2NYkpFDsekEivfJEWcZADaWNrFg7xpEpRLmzIm0nD7HRUVpfJs6At+atfLfjEyvcqCqlTRLA9H5Q3lvxFw0fi9ad/+KpIHU7TE1NRWPxxPWIdcalQJBELrsb+v82JYSmRKtId6gxqhtG3TdVdO2ShsAjuMcW/Cv3XKk7SRpi1QaNAPD3gyMdxEBxk4czo7UfPwffxJpKRHFqtG3b8iRtNNizugMPi2YyehvlqH0+3A53DimTOt+sG0gADZbp12r99cx79AG/NNnQmxsL6k+MfnJ0ZTEZaC0tEJwCLbM4Gbr3gqu2blCqplMS4u0nD7HhaNT2Zg5ElV11Skt1MjInCn1B4+iDvhIGpVPSXyHxdd+FkkD0Gq1pKSkYDvmu7I3SNCrEWg3aW20mYZj0x0DHRZjrN1E2Q7UOwiIUGPtfg7cziobP37vANsre/+99kXkdEcZQGrF/+WQiSg2bRzULZOdrR0+GORI2mkxNjOWfy+8E1eUAfHyK/jDY/+g3ivgP+ecUDopwOpNh7CnZ0lpYsEP9kBAZMPKTQxvKEN9+aWRegvdMjTJ2P5lL3d4lAHE5cuJcdlQ/OQnkZbSJ8mI1dM6aZq0McizNGR6D1EUcRw6DEDsiGF4o2No1Qa74vWzSFobJpOJmJiYsNenHYtKKZBh1pJp7twUSa1UoFEK2D1+Pt3byO9WHSUgipQ1t0fIrC7fsU9HtUUyZ92lSVrdPn7y/kF2VdtD0bjBTsikyemOg5vxWWa+GzEVQRRhEHd5tLVY2jfkSNppoVQIvPfrRWiL96B87VUef/RqHr3v79ToYgj8+A4QRY422mn40V0Yaiolw3PoEADflTZStPkr6Yk6dI3sCxi0KlqHBedg7doVWTEyEafJ7kFbKt0IMmFCZMX0YYrmT2Nb6nC8f36ObQeq8d13H2zdGmlZMgOYeqsbc12VtJGdzej0GOqMUjfC/mrSBEEgKSkJlUoV1vq07njx8mH8+KyumQJGrRKb289L31bx2d4m3tlWR3Ft+2y37pqKVLVKJs3azbHSRleo3b/XH77Uzr6O3e3nqVVHKW10hn6Gck3aIEelVJAwcyoNxjjETz+NtJyIIIoi5RUN7TvkSNppo1EpELKzIToag1bFHdfO4O8TL0OxZzds28bvPtnNOQe+Y1vGCOmCFSsAeHtDGVft+YLA5MkwZEgE30H3JORl0WI0w86d4JNWCcubHDxw53PYx03osSHeMn2fTUeayGytwRcb129v/HqDi4rS+OdZi1AfKUU5exaqF16AZ5+NtCyZAcz+WivplmBGUHY2P5mbR70hmDrfD9Md21AqlaSlpeH1evH7e6/1faJR3W0kx6hRYvf4Q+3hX91Qw4ay9oVuSzdGrNoiGUx7N8c6RuFcvsFp0kRR5P+WH+HTvU18eagl1DhEjqTJMLsghS9yJxBYthx6YYBiX2NDaRNuSzDEfs45MGxYZAUNIGYNS6Dm/AW4lWqaHn0Cz9LlxDotlF9zE2XmFLxLl1HT6qJ6+WqG1R9FcdttkZbcLcNSotkdnw3/+hf+KD3WCy5h8W9e4c8vP4Bh+1ZpQLbMoGB/jZXMlloUfXAxoS+RZo7inEdu5+0x5zKm5hAulUZaCOzlaIDM4GF/jZV0Sx2B2FiIjmbO8CSmTR8VaVk9glarJS0tDbvdHvb5aSfCEIykNTt9JBjUOLwBVuxvZmyaEegaSRNFkSrL8SNpR1tcqBUCJp0St29wNhqqsXpYf0Qyul6/iN3jR6kAraprU5f+iGzSzoDZwxP5Mm+S1Bhh3bpIy+l1dpS3oPN58I0bD6tWgaF/T3bvSwiCwK9+OJPXz7qCuJWf8+/FTwKQdfVCVudOQFi1io9/8QL/evf/CMSY4eqrI6y4e4YnR7MvMQcAt6BAtWolDz5/P3WGWOz6aDhw4NSe8Pe/h5/9rOeFyoSH+np4800IBCipt5Fjq0cxJDfSqvo8V0zKYtLK/7F/1Tp+uuBnCFYrfPllpGXJDFAO1trItTWgyM4O7RNygo+t1siI6kGMRiNJSUlYI/xejBrJpLU4fcwZaiYx2FzkqrFSqcixRqzF6cPplSJD3dWkHW1yk2HWolcrcXkHZySt0d5ex2dx+bB7/Bg0ym47b/ZHZJN2BiSbdNRPnYVPqRqUEQGb20eU143SKJuzcJARq2fa2y/zwGOv0RyXDHPnMnp8Pp9PuRCV28VtLz5GIDkFxXfr+2xKyvmjUhg+dTQAgQsvYuszL+PTG1h626PsSBt+SjPU9tdYcb35Frz+erjkyvQES5bATTfx1ndHWTL/erjhBsR776WkxkJqSy3kyibtZBiaFM3wc6bRPHMuDm0UfPhhpCXJDFD211rJtjdAB5PGo4/CXXfBLbdETlgPEhsbi9lsjkjHxzbiDSoqLR4c3gAJehVXjEmgICmKGUNiEOgaSWtrGgJg68aklbW4yI7VolUJuAdpumOL09fhsR+b2z9g6tFANmlnzJSibL7LHIV/EJo0q8uHwe9BiIqKtJQBy6j0GP781A+Jra+CZctQKgRy589gS8YIFIjE/vff0DYMuw+i16iY+fhPYOFCjC+9yPT7foiqtYXAlVeyPyYV8cCBk54H9ej7OxAPHoTaWmhsDLNymdPhu5IGqu95GF5/nUOvvMEFu1fTGGVCePFF4resR+3zyibtFLlk8hC+zJ2I78Ml0It1NTKDA1EUOVhjIampprNJM5ngxRcHTP1oWyORqKgoHA7HiS8IA6kmbchUmKNU3DgphVevKUClEDBqlV1q0koaXQBolEIXA+fzi1S2usmK1aFTKbqtSXv8s8O8uaU2TO+mb9D280w2qjtF0gYKskk7Q+bkJ/LlkEko9+0bdHNtJJPmBr3+xCfLnBkKBWikYuMnLxlJ3uJ/w7//DTNmRFTWSZGaKkVXUlOlbZWK0ekxHI5LR2GzQVXVCZ8iEBBpOnSUKI/0pUVxcRgFy5wuH/3zI1LLpc6j9//3KbR+Lz+/6B4ALtsa7IIrm7RT4pKiNJYPn4aqoR7Wr4+0HJkBRuOb77LntxeidTk6m7QBSNuga4VCgcvl6vXXT43WhB7H6jvPUYvWKrG5JcOx4aiFRz89zOfFjaSaNAxP0ncyaeuPtHKo0Yk/ADlxOnTqribNHxD5ttTCzqreHUHQ27QExxZkxepocfmwewKySZNpZ0J2LBtGDs65NlaXF73XI5u0XkanVmKeNgluuinSUk6b/JRoDscFZ6h9T8qj6+8v03DrHRxduZaU2vL2A/LctT5HncXF0C8+wafRsmbahRg8Lhy//h3+ixdQYUrkgv3BmX9yg6FTIs6goXTSbCmtfpB9x8iEH/8777ZvDHCTBqBSqcjIyCAQCODxdD8gOlykmjqYtChVp2NGrRKry09Jo5PHPitl7eFWdlTZmTvUHGrdD9Jw6wc/Pswt70jfm0VpRrQqBe5jatLq7V68AZHWbmavDSRanD40SoEUkwaLyyelO2oHjrXpkXciCML5giDsFwThkCAIj/bEc/YXVEoFedPGcjQ+A/GTwZXyaHP70PlcskmTOWVMOjUN6cEbguM0D6lpdeF58GESXv0HKRefxzU7lgMQEATZpPU1HA6+XL+PUbUleMcUMWXFe1R/txX9E49z84xctg0dj9bvw7tgYZ8cFdHXyc5J4UhSljTKQkamBymP7zDPKysrckJ6EY1GQ2ZmJh6PB28vdubuZNL0nU1atFaJzePnuyMWXL4A8/OlEQhzh5qQ6ftBAAAgAElEQVRDx4BOXRzTTBpSTZpu0x0rWqRusIPBpMVGqYjRKWlx+rF7/OjlSFo7giAogb8BFwAjgGsFQRhxps/bnzinMImVuRMQ16yGXp5uH0msLh86j5zuKHN66LKzpOjA0aPdHn/iP99ictlYdeEPqIhJ4rK9qwEoSRsqm7Q+RJ3Fhe2mW5h100IKG44SNWEcOqOe9CljATgrL4FLfnsvJCejfu7PEVbbPxmeEs3uuCwCu3dHWorMAKO1obV9YxBFubVaLZmZmbjdbny+3jEySUYNimDTQXNUV5NmcfmxuPyoFAKPz8viL5cNZWSKAYNGid0tmTCXtz3tcVyG1Lpf251Ja5VMmsU5sOtYW5w+YqJUmKNU+AIi9Tav3DjkGCYDh0RRPCyKogd4B1jYA8/bb5idn8jqoZNQuN1SK/pBgs3tQ+2VTZrM6ZERb6QuJrFbk+b2+anfJt2Qjr32Yu675KHQsZ2xmQT2yCatr/CLN9ahWvIhabVlmJxWGDOm60kXXAA1NXIU7TQZnhLN/oRsFGVl0Np64gtkZE4SR1MrrTHx4POB2RxpOb1KVFQUGRkZOJ3OXhl2rVIKJBk1aJQCenXn2+9orQqb24/F7cOkU6JVKZiYKTVtMWqUWD0+RFHE1SGSNi1b6uqsUylC3R2bHF48vgCVwUiaxe0jEOH5cOGkxenDHKXCpJNMrzcgyjVpx5AOdCgWoSK4b9Bg1mvwT5+BQ6sfVDUDDrtL6tYmmzSZ0yAzTk+ZMRGxG5NWUmcnu0H6WEkYP5qh583kkzueoPTp5zkYn4WiqlK+We0DWF1eTMs/R+frUNsxenTkBA1QCoImDYA9eyIrRmbAYHV58VltEBUFyoFzY3sq6PV6MjIysNvtvWLUUk0aYvWqLnO8jFolVrcUSTNpu9ar+QNSqmNbxOyvlw1l7lDJVGvVCtw+kXqbh4v/325e/LaKymAkLSB2be0/kGgzaWZd+88sJmrg/C6rTnxKzyAIwu3A7QBZAzDvedaoDFZnj+W8zz5DKYowQAbpfR9eWzC1U27BL3MaZMZFUWFKZNKRfRz7kVpcbWFIUxWiQoGQl8cLI7RwzThcXj+Ht/w6eFIxTJ3a67pl2ll7oIF5+9dRH5uEX6EkpbFaNmlhIDNWT3l6nrSxezecdZb02GaTbq7lz2CZU8HthuZmip0a9F4XymhjpBVFFIPBQEZGBhUVFRgMBpRhNKwXFsZRZ+taB2fSKXH5AjTYvZh0nV/fqJW2rW5/aGh1hlkbMno6lYDLG+Bv30qdkteWtHR6jlaXLxRpGmi0mbSOxmxKdt+cG3s69EQkrRLI7LCdEdzXCVEUXxFFcaIoihMTExN74GX7FucUJvFV3iSUVVWwY0ek5YQdjy+AwuWUNuRImsxpkBWnp9KUiKKmGo4p3i6utpDXUgU5OaDVhvbr1EqEkSOlDbkuLeJ8sa+WcTUHiT//HJJv+QGMHTvoUqZ6A4VCYMiEQmxaPeLmze0H5s+HhYOqukCmJ3j+eRgzhuKqVqK8brQxA2MW2plgNBrJzMzE4XCEtUbtohHx3Dw5pcv+eIPUkr+00dXFpBk00q263eMPRdJ0qvbbd51KgTcgsupAMyAZl9JGF/mJ0uJN6wCtS/P4Aji8AcmkdTChefEDZ9GqJ0zaJmCYIAi5giBogGuAj3vgefsVw5KMFI8LzqwaBIOtbW4fOq8UTpdNmszpkBmrp9KUhBAIQGXndZ2yw1WMrzuEkJ/f9bpxhbhUGvy75bSvSHNk1yGSLQ0oJk9G+P3vYdOmSEsasJxdmMy6rDF4ly4HUeSDz7fAd9/BypXwxReRlifTnygvh/p6Wuub0PvcqEyySQMpopaZmYnL5erVro8ACUGTZvP4u0S9okORNB/OYOMQXYeaNm3QsAVEGJViwOMX8YuSIYSB2+Gx7X2ZdSrignPnZufFRFJSj3PGJk0URR9wN7AcKAYWi6I46O6eBEFg/OQCdqblExgEJs3q8hIlmzSZMyAzTk9rUnDAdYe6tDfWH+Gep+4gubkWbr65y3XjhyRQEpeBfbvcjjyStDq8xBYH/x9MmiQNXFcNzJSavsDc4UmsHjIBTUUZrVt38vUriwGwq3X4/u/JCKuT6VfYbACIdfUYfG4E+Ts8hF6vD3V97M05am0mDdpNWRttUbZ6mxe3T0QANMr2kpqOUbVpOVKqX5xexdRsyXxbgmZGHGANROrtkpGO06swapW8cV0Bvzk/N8KqepYemZMmiuLnoijmi6KYJ4ri73riOfsjZxck8UXuRISNG6G+PtJyworV5SPKJ5s0mdNHqRBILyoAIHC4FACHx8dH//iAUbUleJ5/Aa66qst1E7Jj2ZE6DOPar+Dvf+9VzTLt7KpsZUz1QUSFQkpzlAkrSSYd1dNmA7D/34uZUrodT7SJP8/8Aap138LGjRFWKNNvsFoBEBsaMHhdYDBEWFDfIioqiuzsbHw+Hy6Xq1des6NJOzaSlhItzVers3lxev3o1IpOjUe0HaJqY1INxOtVzMkzh9r8tzj9+AMid/zvIL9cfiSM76J3aZsFl26WSiLyEqJQKQdWP4iBM5a7DzB1SDzfFExFEEVYtizScsKK1eWTI2kyZ8yoaaOp15vx//wXcOQIm480c9Hu1QQ0WqJuuL7ba5JNOl679C52F0xEvP9+aG7uZdUyADsqWhhfuY/AiJHyTV4vMXraGHYn55G0+L+cf3gj6nPns3X+FVh1Bnw//wV07E63Zw989FHkxMr0XYKRNKGxEb3PLf/9doNWqw01uXM6nWF/vRidElVwiNqxNWnRWiU6lYJaqwe3T+wUOYPOkbRYvYrXri3gpzPTMWqUKAUpLfDd7XXsqrazYv/A+b6sbJUinekm7QnO7L/IJq0H0amVxE2fQkN0HOIAT3m0uX3tbbdlkyZzmswYlc6NV/8aoaEenn+edQfruXjf1wQuughijp9bPn5UNj+fcA2CxwPvvdeLimXa2HOwiikVe1Ced26kpQwa5hYk8dqEBeTUHcVsa0F48EEeu3oyz868AdWqlfDoowC88fVBGDUKLr20S1MeGZm2SJqqqRGdRzZpx0Oj0ZCVlYVKpcJut4f1tQRBCEXTjo2kCYJAUrSaWmt7JK0jHU2aOUpFgkGNViVF20w6FRaXj5UdzFmTY2B8JlS2ukkwqLv8PAYSA/edRYizR6bwRc4EAsuWD+gvR295BdfuCEYLZZMmc5okReswTp7A7uyRsHYtVd9sJNHegmrBJd973W8uHUXhxXMoic9AfPPNXlIr00ZVixPvyi/Q+L3SoGqZXqEow8zGqefSGJ+CeOmlMG0ak3PjEO6+m/dHn0Pgr3/FWlVLxVPPtV8kz1WTOQZnszRjUtPShM7jlE3a96BSqcjMzMRoNGK1WsNa1xVvkMxZjK7rCIBko4Y6mweXt2skTauSInACdOpyCJJpa3H6aHL4QmmTB+udeP2BMLyD3qWy1U1GzMCNooFs0nqcucOT+DJvEkpLK3z7baTlhI2CP/2G8w+slzbkGT0yZ8AlRWl8lVyIuH07Q75eIe08++zvvUajUpCfYmJ17gTYsqUXVMp05D/rSrl47xoCBiPMmBFpOYMGhULgowfOxli8C+Htt0P775o7lDemXo7C7WbXU3/l0i1LaYwKzgrqTx03S0pggDU36HX+9Cd47bXvPaWhuhEAfWMdKr9PXmg9AQqFgpSUFOLj47FarWEbep14nEgaQHK0mjqbB7cv0DWSFtw26ZQoFZ1rsuINKurtXpqdPiZlSY1E7v+ohLP/voPfrjxKf6ay1U26WRNpGWFFNmk9TEqMjqazZuFVqQdsK/6KZgdNh8vad8gf8DJnwIWjU9mSPQpBFPnRxg/xDcmDkxh4nxitpd4Qi+BwQJhTUWTaOdpoJ/2Xj7Fw7xoUt9zcaY6dTPiJM2jQJiaAThfalxitZcplc9mWNpzh//4bhfVH+PSiH2KJiu4XJm1/jZV//PAXMHQo/OUvkZbTv3nlFenfcWh1ejF4pBorc321tFOOpJ0QQRBISEggLS0Nh8MRlhb9oXRHbddIWpJRQ6Pdh8Xt6yaSJm23NQrpSLxezdEmN76AyJC49s8Mc5SK5fubcPv6Z0TN6fXTYPfJkTSZU2daUQ7fZo3Bv2TJgFwV/M+6I2TWBk2aRgNxcZEVJNOvSTBqueIni3CotRi8LlQnWeOUGK2lUR+sW6urC6NCmY489/Z6rt3yOY7rb4QXXoi0HJkgP56Vx/PzbiHe3gKAYtEiticPxbehb3d9FEWR3y/eyOX/+xt+QYH3wYex3nl3qG5K5hRpbpYiksdh1d7akElLaw12oZZN2kljMplCnR97uqFIikmDUuiasghSh0cRqaPh8WrSujNpCUY1No8U+YvVqxgSLxm1e2Zl4A/AgXpHj76H3mJ7pdT8JtMsmzSZU+TswmRW5U1GWVIC+/d3PcHhgDBOtA83R/aWkmxrlNIqXC453VHmjLls+jBq16zH8/6H8NvfntQ1SdFaGgxmaUM2ab1Ci8ND/JLFqAM+9I8+DMLAanfcn4kzaFh473Vsv/Z2uPpq8ieNYFvacJS7d4HFEjrP6w9w8L3PEEeNguLiCCqW2FjaxNAP3yLR3sKvbnua1bnjiX75b/D555GW1v8QRWhqkkYAtbZ2e8rWklq0fikKlG4Jfm7KJu2U0Ol0ZGdno9FoerRObeGoBF68YhiG7iJp0VKUrdXlP253x9juTFqH1v5xejUvXj6Mj28dxfh0IwC7q/ufSXN6/fzxq3IyzdrQXLiBimzSwsCY9Bi2FU2XNj7+uJsTxsCzz/auqB4iEBBhR3CA7dix8k2aTI+RO20cmssvPenIbKJRJ0fSeplVxXUs2LMaR9F4qXugTJ/i8vEZjH3rH/DOO4zJMLMpewxCINCpPvo/645Q+vPfIuzZQ+CKRdJCWwTZcrieH275BN+s2Tzx0kM8tOAh6cDR/l0vExFstvYxDMeJpjXUNIUeJ9mDHf9kk3bKqFQqMjIyerROzaBRUpRm7PZYcnR77VWUuvt0x1i9mmNJ6LAvVq8KdX+MN6hJidawp6b/lQrsq3NSa/Vy1/Q0otRdDe1AQjZpYUChEBgxeRTFKXmIHx1j0vx+6cNz167IiDtDjjTayak8JG0UFUVWjMygxhSlwhIdK23IJq1XWL3pIKNqS4haeHGkpcicgCiNEteESfiUKli9OrR/w6b9zC3ZxI6UYSiK90Z8llpg5SrSrfWoHrgftVJBbEo8Dn00lJWd+GKZTry7Ykf7xnFMWkutZNLcyg5RF7mu/LRoq1PLyMjA5XKFdZ5asrHdpGmPjaSpvyeSZuwQSTvm+MgUPcV1/S+SZnFJmWgdjetARTZpYeLsgiSW502G79ZLqQdttKWdVFdHRtgZsrvKQm5zJb74BIiPj7QcmUGMIAgIycnShmzSwk6txYX9q69RigGEWbMiLUfmJBg1LJXtacPxf/UVIM23TFq1FHXAz5u3PE6dKYHA669HVKO7pFR6MGECADkJBqrNyf0vknbllfDggxGVsOSL3aHHzr1dSy0CARFLg1SzWGlOaT8gR9LOCKPRSE5ODiqVCpvNFpY2/Tq1IjTkurvGIb88L5uFo7rek8UHI2kKAWKOMWlZsTpqrR58/v7VO8HqlqKWxw79HojIJi1MzBiWwOr8KVKqScfc+hbpA5KqqsgIO0P2VlnIbalGMWxYpKXIyGCKM+HU6aG2NtJSBjxfv7aEO9a/h6hSwdSpkZYjcxKcPyqFtdljUWzeDCUlrDvUwJQjO3AnpzDv2vP4oHA2wvLlnRcSexGb24fY9tqJiQDkJhg4aohH7G8m7X//gz//OWIv7/EFoKV9YHH5x8vhJz+RaseDNNjcaJ1SelvxkNHtF8sm7YxRq9VkZmYSFxeH1WoNS/fHtmjasemOAOcOjyPR2DWy1FaTZo5SoTimPCXVpCEgQo3V0+Naw0lbJM2k7Ro5HGjIJi1MROvUGKdNocGU0KkubceuI9KDfmrSShts5LXWoBg2NNJSZGRINGppNsbKkbQwU9ns4NIHrmdK+W6EggL5pq6fMDknji+nX4JfoYTnnmP1/jqmle9CNXcuM/MTWTpyDoLfD598EhF9+2ssxNlb8RmMoVEOuQkGyqITEftzuuPzz0v/epFaiwuTU+p415KYSv7Wb+Cll+CJJ0LnlDc7Q50d942Z1n6x/PfcIygUChISEkLdHx2Onk0lTA42Dzk23fH70KkVRGuVxOm7Gpo0k2Tqqi3unhHYS1jdfpQC6DUD38IM/HcYQc4uTGbZkEkEli8Hlwt/QOTZdzdIB61Wqci3n1FZ3UxCaz3k5UVaiowMySYtdToTomzSwspTf1+GKhDAPbwQ/va3SMuROUkUCoFpM0fz0cg5iK++StXq9STamlGePRe9RoVx6gTqzYkRM2kl9XbinK2IwSgaQE68gSpTIorW1uN2KOzLOKvr4P77pX+9SGWLE7NLGlsgrFrF1T/+G1/OulRqDBOslapodoRMmiu3w3e4bNJ6lKioKHJycoiOjsZiseDroW7eSd8TSfs+Egxq4rppKpJqkhZGqixSJO2zvY1c88Ze/IG+nf5odfkxapUIg6BxnWzSwsg5hUmsGjoFhd0Oq1fTaHejd3aY/dLP6tICARH/4cMoRFEaOiojE2Em58ZRFxWDq7Ka75Z+y0dX3IHvSD9Lk+rjOD1+hE3SrC3tf98AuR6tX7GgKJ0lBbMQnE5u/Oz/STtnzwZgVn4Sy3InEVixIiJdHuutbuLtrSiTOpu0SlOStNEPo2lL7/hFRF63utVJjEta+I3JzeTsa89jScwwqS1/SQmIIlX1lpBJ08eZ2y+WG4f0OEqlkuTkZDIzM/F4PD0SVWtrlHFsTdqJeHBOBj+eltZlf5JRjVIB1UGTtvJAM2XNbhrtPZ+q2ZNY3D5M3cySG4jIJi2MZMcbqJ04DZc2Cj7+mAarh2h3hz/UfpbyWGNxkdZQKW3IJk2mDzBzWCINBjNidTXunz3Gwg/+ASNHwqFDkZY2YNhfa2V09UH8Gi2MHn3iC2T6FKPSTdSOnYJDreXsw5vxjBoNwZriWfmJfJE3CYXDAd98I10Q7hmeHg8cPAhIJi3RZUGRlBQ6nGrWUR0TNG39pS4tEAg9XPDJv6QHxu5bqYeF555DvWwpMS6bVDNqNHLDtGwa07Ol4wcOwA9+wJ3njiBJId2AZ2Un8YsrH0McORJMA3vWVCQxGAzk5ORgNBqxWCxnVKsWSnc8xUja+IxoCpO7GnGlQiDZqOH1zbXc+u5+dlRJJr+v1aj5A2KnodtWl5/obmbJDURkkxZmZo/JYk32WAIffUyD1YXJ1WEmRT8zaUca7GQ310gbcrqjTB8gzqChdswk9K3NzN79NZ8VzsQrArfeKq0gy5wxxdUWxlYfwDd6DGgGfsvjgYYgCCyanseG7DEAaO69JzTfsiAlmiOF4wkICli7VrqZj46Gr7/uWRGvvw5z5kh/k6+8Is3Ya26mweYm3mkJNQ0BUCsV+DOzpI2yMnxeHzz3HJSX96ymnqRDFFIlBg1bx5ll4eaBB7j48ds5v2QDQmwsCAJ6jYr4sSOl4wcPwltvATDBJmXwXDazgCfe+g3C7t2gHBw3vJFCpVKRkpJCZmYmPp8Pu91+Wh0g29ITjT1Yi6VUSJ8FxbUO3D5JU1tk7VQRRZFfLj/CxjJLj+kD+HBXAz98ez9bK6RMNKtbNmkyPcS8wiRWDp2CoqoS3+YtmNztdWhiPzNppY12xlbvx5+UJLffl+kzZNx5C6VJ0opxxR338efp10k3nMXFEVY2MCiutlBQfwTNxPGRliJzmvxo5hBmPPUzmDgRrrsutF8QBCaOzmZv6lBcq77i9Yf+JBmOJUt6VsDXX8OaNVKK/65dUjRt717qLS5i7S3QIZIGYMhMx6tS4zh4mN9e9iA88AA89VTPaupBvtnZOS3z/SkLpAe9UFNXUtpeNpFXXwaxsaFtldlMozEWDhxAzJKM7+Q960AQUBgNaFWD40a3r9AWVYuNjcVms+F2n1rDjlEpen57QQ6Tsnou8plhlozf6FQDyqAjONTg5KtDLaf8XC1OHyv2N7PhqPXEJ58CVcHGJhvLpOe1uPxyuqNMzzAuK5ato88iIAiYVi3D5LLj0RtwqrS0HDgcaXmnxMHKFmYd2YbivPNDK7EyMpHmyqk55H78Lvz618y5aj4rcydKB+67T+oY19gYWYH9nPJDFcS4bAjDh0daisxpIggC6isXwaZNXeqPZucnsj59BIpNGxm38QsAxC++6FkBbW32d+5sT0XeuxdHQxMqv69TJA0gM8FIjSmR1q07uPeL1wDwv/MunOJNbW/x0tL2+WSOuAS+TQhmmjQ1hf213/xQakZWEpch7bC23yAbtCqOxqfDgQM4M6SFLHNdlbTIKn+HRwSlUklCQgI5OTkoFAosFgv+k4y4CoLA2cNiUSl67v/dE/OzeeO6Av5w8RBevHwY5igVb22t4+efl1JvO7WIWlsEzubu2QiyLxic3l0tZaJZ3T45kibTMygVAuMn5LMjo5C0r1dh9jpQxsexI3UYgRUroaKifXZaH0YMBKhauRaz04pw4QWRliMj05kpU+CJJxieEk3MmBE0GONg5UppxX7Llkir67eIooir+IC0IdehDkjOHZlMychJaHweRteW0BhlQtixAxoaupxbZ3Hhv/QyKbJ1krU1DTY39gopa6Ru3WZadgUj3MXFBOo6z0hrIzMuinJjAvHrvybWZeXtSZegbGkmcONNUFp6+m82TDgswQyZl1+m5LsdtOiipe3m5uNf1AMEAiL7t0t/n0dvvF3a2aEhmUGrokYfC/X1uCwdukk/80xYdcmcGK1WS1ZWFmlpabjd7tNOgTxTzFEq8hKiMEepKEozkhKtoU3F0eZTWxRp6xJpdfdsXWtD0CzurLZj9/ildMdBMMgaZJPWK8wrTGb5kMmklexlRGMZythYdk2ZR/yRg5CZCTfcEGmJ38+TT+JLS2fs9rUEFAo499xIK5KROS63zMjlu/QR7TuCTQpkTp2KZieJdcFaILkOdUCiVSmZ99At/G7uLViLJvDPRfcB4H36953qOr3+AFf+/F2UHy2RasR+/euTev6nP99H3eEKAMqXrcZUL9U1u3fuRtscjHJ3MWl6Kk1JaLzSTWLck4/x/si5+JZ8ROCKK8Lf3OQUEe3BpgaJiRQMScFp7B2TVm9zE9cq/QzPvvUyuOAC+MMfQseNWiVWlQ7RZsNlsXEkNg3/f/4DP/xhWHXJnByCIGAymcjNzQ2lQDqD4xIiRUp0e91xRcupmjTpfEsPR9LqbNKCkC8gsrPKRkAcHIOsQTZpvcLMYQmsKZgKQEHlATCbUV95RfsJfX2l/9e/Rl1bw62bluCbdy7ExUVakYzMcVk4Np35916PXwh+vMm1aafN3moLOc3ViIIAQ4ZEWo5MmJg3MoW7P3uZ6O2bmfnE3bxVdD7qP/8JPv00dM6hOhtFR/cA4FOq4NtvOz2Hy+vnhYf+QkXGUAJjx4XSGiuaHcTbpdqsUVvXoEDEpdIQ2LuXOEewwcAxJi0rTk+VSdrnMZk594IpHH3uZe678D4U27YhPvZYn2kM5A+ICG031no9aqWCpKxUaTvM6Y7lTQ6SbEEjmJoKn38OP/tZ6Lheo8Kh0eFuseC1WGkdPQ7ljTeGVZPMqdMxBVKn02GxWPB4ItNhMdXUbtLKWk5tLEdbuqPV1bMmrd7mZVSKNMvvQL30tyZH0mR6DINWRfLksZTGBj+4zWZmzCnipSmLpO1j298+9lifWelyef3UmhIA0Pp9aB7s3QGdMjKng/b2H3HfL9/mQFYh7NsXaTn9D1HEf9ddOJd8TE5zFWJGBuh0kVYlE0ZioqT23lPzEnj6wrvwqzobsT1VFsZXFuPURvHJqLmIu3d3uv6ZV7/klr8+itvpwllSCpdcAqKIpdWGyePArdKg9UsRsG+yi4iqriSvKRilTes8w2lMhpkxM4LdKMcVISgUPDA/n/EP/Ig3x16A8OyzUpfIM8HphNdeg7ffPiPD5/D40PmCN9RRUdJ/2xYywxxJq2h2kmRvQtRoul08NWpV2DVRqJ0Oov1uxgzvOitLpu+g1WpJT08nK9jkpScHYZ8sU7OjmZQZTXaslvJmN25fgC8PNhM4ib+RqtZgTZqne5P2/o56lhafWo24LyDS6PAyOvUYkybXpMn0JPNGpPBF3mRpw2wmL9HIe4vuYvk5V3Udar1hg9QJqw+wZEMpCdYmfDFmuPBCmDcv0pJkZE6MUknutLHsjk4hUCybtFOlfud+lH//O5f+4sdcvucrFHI92qBBqRDIz4yjPDlH6sQYZE9VK5OrinGOm8jOhByE+nqorQWkxbwRLz2DjgBfPvsqv5l+A+zbh7hjB85KKb3R9bNHQ8/1xdApAJxzaCN+vUGKAh2jYe78YAOgoqLQ/ltn5PLezY+yL7MA8eWXpZ0Oh1R72h2/+hWcd16X3e9sLOOdmYvgllukbpebN5/Sz6gjDo8frS+YFhY0aYGY4KDoMJu08iYHibYmSEnpthGIIWjSlAE/JocVwWAIqx6ZnkGv15OdnU16ejo+nw+r1XrSzUXOlElZJl64bCi5cVGUt7r5dE8jv1h6hPe215/w2u+LpHn9Af60poLfrDy1AfXNDi8BETLNWmKjVOwKNg+J08vpjjI9yLzCZFYNk76YiIlBEATmj0xmpz8KLBawd5ifZrFAXV1khB7DVys2oRQDKJ9/Dj77DBTyr4xM/+DC0Skcis9EUVXJ7i838OmE82j843OdBs/KdCUQEPnwuTc77ywsjIwYmYgwOj2GXbGZiB1MWtWeEobXlmI67xzK04Kpr8Fo2vrDjYwr20PL3HO5/vqz2TZhNn5BgevtxRgsklGJmTgW9u6F559nf6AqQjMAACAASURBVP5YACZWFBPIy+u+0+Dw4dL3zdSpoV2CIHD9WbksHjYTYft2aa7b7Nlw551dLm+ye/C89TasWAF79oT2e3wBFi9ew6Jty/m4cBZ+hRI++ui0f1Z2t48ob2eTpozS4VJrw5/u2Owgw9WKcIzJbcOgVWJXSxFwld8HsknrNwiCQHR0NDk5OaSkpOB2u7HZbAR66fsrK1ZLZaubertUD/bmllo8vuO/ti8gUmP1ICBF0vwBKfJ2qMFJg93Ljir7ca/9Ptrq0RKNalJNGhrsXgQgLz7qtJ6vvyHfcfcSKTE6fvp/N+MbMRLGjQPg3BHJ1OiDKQodo2lWq7Q6aLN180y9x+F6G659UtMFQV5Jl+lnFKSYECdNAiDt0gu5eOsK4h95ABYvjrCyPkxpKW+u3k/C5nU4Y+P51Wtr2f6vxSfdJEJmYDAyzcSe+GyE8nJobsbnD1CwcglKMYDqB9djnix9h7WZuK92V5LZWkds0Qj0GhWLzhvPhsxRBN56iyRb0KgkJkpm/957IS8Pr0KJAhF1wXFGO2RmSvWkV1/dafclY9JYPWY2AL7XXkPcskX6mz6m4cL/vboGzYH9ADif/BV8/DEAy1Zt45nXHkfQaSl55FdszBiB9/0PTvtnZXf729Mdg+MNdGolrVHRvZLumOpo7hKJbMOoVeHQdLiZPWb8gkzfR6FQEBMTw5AhQ0hMTMTlcvWKWcuK1eIPwJbgAOlGh491R44/pHpHpQ1fQGRMmrQQYPf4+e6ohRvf2sfTq8r4plSqSz3VNMX6Y0waQHasDr1GTneU6WGmF6Sg2rMbbr4ZgHGZsbgSggXTx5o0CKWSRIqPtleR3RLUJXd2k+mHnH/P9Xw6ei5x1iaemXszLaY4+OD0b8gGNI2NMGQI15w7mvNKN6Obfw5P/nAmY2+5Uh5eP8gYnRHDvsQcaWP3btbsr2PB9pU0T5gCQ4cyZnw+9Xoz7ncXg8fDoS37UAd8qPKHAbBwbBqvTlqIoeIo9377tvQ8HQZWpyaaKI1NlzaGDTu+kPz8LtkbURoluUX5HEgbivXFlxFEUVrQXL48dE6r04tjzTcANOpjiHr/PVi4ED78EPcLfyGnpQrF0qXMnD2GlUOnot5XDEePtr9IICA1TTmJeiB7NzVpOrUSi84YdpNW1WAhqaWuS01fG3qNKhRJA+RIWj9GoVAQGxtLbm5uJ7MWrjTI3Djpd3lvjYNRKQbMOhX/v737jo+ruvM+/jnTNRr1LqvLvVeMjTHFFGNagABhQ0jbhRQ2Ibsp5EkPSZYNKbuBNJLNk4SHTSFAAqEEiDGh2IAbbrj3bku2uqZo7vPHHcuSLdmSVWY0+r5fL708t8y9Z36yRvrNOed3Fm/t/v/zP7bX4XEaLhttL6beEGzju4vtOaebjzTz5i47wQu39W7+54khlIVpHgpjSdqY/OHRiwZK0uLK4TB4S2MLUMaStIbWMFaCJGmrlm/i2kPr7E/fCgvj2haRczGlLIsLX36S9Q/+itZP3cPz1bOxnn0WWntXtSrZBSNtfPM//gCApy2Ct6wE85GPxLlVEi+j89M4UDHa3li5kudeXkN17V7S33czAHNH5vEfl3wY37KlhL5wL44d2+1zYyMu8tN9BBcu4tXyqUw9EFtnr0MFx9IsP1tyS+2NMyVp3ZhTncOygjFkNdprjEacLnjiCTYdbODgAz9i99e+w6y9G4i63TQ88zyfu+leNpeNpe2f/5nczeupKx+JmT+fkfkBVhfFXueaNYA9HHLlAz+zC5/8+tdnbYtdOKTzcEevy8Fxb+qAJmnNoQhVq97A19LU5bw7UE9aMnI6ne3JWkFBAaFQiIaGhn4vMFKZ7cMAFvZ8sPnVGby+o55gF0MeLcviH9uPc15ZOvkBO5Gqa420L4Zd2xxh17EgKW4HrZHoGYdNnmpvXZA0r5N0n4ui2PIAY/OHz/9jJWlx5iqJfQIWS9Ku/q9XMCeGOcZxXlpTMMKtv/0us9a8Bpdc0vWcAZEhICMngwl3f5j544t4duT5mKYmDXk8xcYDDYTW2vN2jqzfjOvdDd3+4SfJz+EwjJsxlr1ZhUQWv8z+pasAcE6aCEB1XipvzruGV6YvwPGrXzHqSKwXqsOIiy9dM56vXNFhrlhmZvvDqaWZ7CkotzdGj+51+86vymHFiLEAhDw+FlfPIvrqa3zzv58m595/Z8L3vsEHVj+LuehiKi4+n6mf+xg/mXw1ztpaLtj5Dv4Z9py4TL+H2vJYm2Pz1n7z+g7cP/yBve/3vz9rW5qCbafNSfO6ndR5/Fh1db1+bT21Zm8d161fQjgjCxYu7PKcVK+TJo960pKR0+kkIyODyspKioqKiEQi/Vq63+d2MCLDC9hl+S8bnUVLOMor246fdu7RpjCHGsKcV5ZGIDac8UBdiDYLpo0ItC+OPbfCrmTe0It11PYeD1KSabejMjYPbVLx8Pl/rCQtzgLFhYQdrvYkreZQh0/e4tiTtnL3McqPHeDYBRfB00/HrR0i/eXCkbn4r17ImsJRhD9/7+lzPvfvh3A4Po2Lsy2HG6mu2UM0NUDeOM0/FbhoTB6vlU6m7eUlVB2OJWGxAjLGGL5yzXh+OfIiXPV1fHj5U1heb6dhd2ML03ng/7yXZd/8b7jrrk4f9C2cWMi/fPsTduXGSZN63bbxRensHmNXfQyNHsPbxeNw7NzBXb//HhGHk9qUDFJMFPPjhwC4elIRa0rstnuiEXzTp7Zfq6CskJrMPFi/nobWMG/++gkmHdrG9qxirJdfhoMHz9iWpmAEbyRkryXotf+Y9LoctLq8WM0DtzDxqp21XLb1LaI33AAeT5fnpHpdNLvVk5bMHA5H+4LYpaV273RDQwOt/TBapCrHTvCLMzxMLwkwIsPDk2uPnnbeiSGJJRle0mNJ2p46+4OLmaX2wu5OA7Nij3uTpO2rC1ISSxanjQjwxzvGM75ASZoMkoIMH4dTswjv3QdAaqjDm3ock7S3dtRSVH+EwJiR6kWTpOBwGB64dRr3X/Vx3IcOwP33tx+zduyAESPgvvvi2ML42XKogTE1ezDjx+vnXQC4cFQeS8sm4W2o47p3XyGammoX84hZOLGQ3Ouv4kAgh7K6Q5hA4LT5Y7Mqsjn/K5+CE+XyO3DOuwBWr4a0tF63zeEw/PJrNxOtqCBl3hw2Vk4AYP7OVRy57YMcevKvOBcvbu+ly/R7KJ81kaP+DPsCHRLDkfkBNuaUYa1fz3NrD3Lbq48Rys7h36/9d0w02mmuW1eaQrHCISkp7T87PreToNtzWjGT/rRjzRYCoRa8s2d1e47X5SDo7ZCkqSctaRljSE1NpaKigrKyMnw+Hw0NDTQ1NZ1zkZETSVpRuheHMdwwKZd39jexo6bz/+v2eWPpnvbCIHuO2UnauAI/freD6tyU9qGQDcGeDc0Mt0U52BBiRMbJDyFO9KoNF0rS4qwww8eurCLa1r8LQD4duqrjmKRt3HGInJZ63BXlcWuDSH9L87kpu3oBT028hLYHvkfo579g675a/nLrv9onnOUPsmS15XAjo4/txYxXqX2x5Qa8HD9/HlEM5+3dgBk79rQE/uvvmcKXbv8Gx7Py4OabB7V9OWk+HMuW4XzgARZ96GpCDnvdpLKvfo4JV1wAc+Z0Ov9zC8fRMiOW0JySpG3ItitJrn1pKQu2vY37X+8mOGUaLV4/vP32GdvRHIzNSUs5mQzZPWkDl6RZlkXdOvtvhjMV9TLGEA10SMzUkzYspKSkUFxcTGVlJVlZWbS0tNDY2NjreWuzytLJS3VTHUvWLh1lFwVZubfzKJQDDfbfrUXpHtJ8dpK2+7jdk5ftd/FP0/O5dWpe+7H6LtZR68rBhhBRa/glZh0pSYuz/DQfawpH4lm/lnBL68n5aBDftdL27LX/7fDJqUgyeP/scr5z0QfZGcjF87E7OXjldSxaEUvOEmR9wsF2cPtecuuOaj006WTyrHH8ddyFAJguKnxm+N1873v/gnP/PvjJTwa7eVBQAIEAt80fQ3jehTRfc3178ZJTjS9Op/TjH4H586GsrH3/3OpcNueWYVpbmfPIg1jGYO66i5nVeawrqCL6dhcLXTc3w+LFcOwYTaE2UiMhTIckzed20uryDliSVtsUImNfbFHgsyyP4wh06KlUT9qw4vF4yM3NpaqqisLCQtra2mhoaKClpQXLOnuVxWkjAvzloxNJ99kfgBQE3OSmulh3sPOaZwfqQ2T7XXhdDnwuBy6HYXesJy0rxcVHZhdx1bic9l62ng533HvcvsaJ4Y7DkZK0OCtI97K2cBSOcIjGFasJxIY7RoyD6FnGwg8kz8FY6X0laZJkJpVk8MS3buFHDz3FI9MWMW/967SmZ/Lz826EnTsHfAHaRNMcijDp7ZftjQUL4tsYSSgXjc7jh/Peb2/Mnt3lOdmpHtJ87rgPk0196W/4n3jszCfddhu88kqnto4pTOPIvEsJOVwsevc1GmbMhqIizqvMZlXBKHhndae5quFQmIZxk+yfle98h6ZghEA03Kknzee2e9JM68AkaduPNlFxfD9Rt/usv6Od6kkb9pxOJ+np6VRUVFBeXk5qaipNTU297l0zxjCxMPW0JO1gfai98qIxhhEZnvZELCOW4AFnTNIsyzotcVy+pxGDXV1yuFKSFmcF6T7WFNkliMPL3mqfk7Y3o4DIvgNneuqAsSyL1MOxe3f4xFEkWRRnpvD5q8bx+xs/we4rrsPx2B95rWq6fXDVqvg2bpC9vfMYN6x/mabKkTBjRrybIwlkamkm8xbNZetba+HLX453c87M7ba/zsFVl0/jmbHzAEi7zR62OaM8i7WFI3EEg+2VH6NRiy9+8/ek7Y4tObBhA02hCP5oqFMC5HU57SQtEunRWmu9tf1II+XHDtBWVgEu1xnP9ad4aD6xVpqStGHNGIPP56OwsJCqqioKCgp63bs2oTCVfXUhjjWf/OBif32wfQ0zsMv3A6R6HHhcJ9OMNK/9f7Wh9fSfif9deZjbH93Yvn2kMcTja45w5dgssvzn9nOdDJSkxVmq18Xx/BG0BNIxK1eSGmoGYGtOCc4D+6EHPzT9rTEYIe94bNhXScmg319kMIzITOGZL19N2d/+QuCyS2gaH5unsnJlfBs2yJa+vo7z96zDc8ftce8NkcTicjq47z0TGTlrYrcVBJPBLTNLmf7T72JdcAHmttsAKEz3sbMyNvx3uT3kcePBBjzL3wRgS/FI2LaN5mAb/rauetJin/4PwJDH7UeaqDx+ANeos1diDXhdtJwoHqLhjhLjcrnaS/iXl5eTlpZGc3MzDQ0NBIPBbp83rsBO9Dcfsf9ft0UtDjWEKeqQpJXHkrTMlM4fILicBr/b0WVP2oq9DeyobW1fQ+3JtUeJRC0+Oruoby90iOtTkmaMudkYs94YEzXGzOyvRg03hZkpbCsfi/edVaTFetK25pbhbG2Jy9Cr2qYQxQ1HCGZmd/rFI5LMRows5UBmwbBK0izL4tCSpQC4F1wa59aIxIcxhvL5szCvvQZFRe37AuPH0pgSaC8esnR7DdP2baIpPYu/l07B2rGD5tYg/tMKhzhpdcf+aO3PJO2JJ6Cujm2HGyk/fhDTgyStJCuFkD+WnKknTbrg8/nIz8+nurqakpIS3G43jY2NNDY2Ej5lWZoTiVdjLNE63BgmErUo7jBvrLKbJA0gzeekIdjGjpoW1h04OWxye41daORYSwTLsli89TjTRgTa12obrvrak7YOuBH4Rz+0Zdi6YGQur2eUk7r5XbKa7cUvt+XGhhnu3Tvo7TnaGKKgoYZQ4fD+BEOGl7GFabyTX0V0xYp4N2XQbDhQT+GuzfbGOaxXJZLMxo/I4J3CkVixnrSl22o47/BmwrNmszOzCBMK4Tt04GQJ/pgTc9KA/kvS1q6Fm26CH/yA+m07SQ02w5gxZ33a16+bQEFRDjidSd0bKn3ncDhITU2lpKSkfTikZVntpfwjkQh+jz2vrDls93htj5Xjr8o+uWh6RexxVhdJWrrXRX1rG+9/dCN3PrbZvn4wwuFGOxmsaQ6zvbaV3ceCXDIyc0Bf71DQpyTNsqx3Lcva1F+NGa6unFDIqvyROCNhZu57F8sY6sur7IObNg16xbnaphAZrU2Y7OxBva9IPI0tTGNdQTWOLVugvj7ezRkUS7fVMObITtpKSiBTvxBFOhpXlM47BSNh7VrCTc1MeOSnlB/eTeDKBRzItRfuzjqwB184eNqctGBXwx137ICnn+51O9qiFsseegSA6NN/JXVLbO7OhAlnfa7P7cSRlma3T8OZpYdODIesqKigsrKS3NxcIpEIbUF7Sk5jq51UbTuRpOWc/JCiPMuHofuetPoOc9J21rays/bkwtu1TRHe3GX//p1frd9JmpOWAGZVZLE7Nvb9/N1rIBDAcaJgx6232mWGB1FNY5C0YBPOLP2AyPAxtiiddQWxNYfeeSe+jTlHbVGL5lDPCxVsWLmZybW7cU6ZMoCtEhmaxhels6ZwFCYcZsUv/8hn/v4rDl15La57Pk3xjIkAzF6xmLwj+6CwsP153fakXX45XHcdNHWujnc2SzYdhueeA8CxaiXn71htH+hBkgZAIKD5aHLOPB4PWVlZVFVVMW6k3YHQGIzQ2NjI5kON5AfcBGKVG8FeJ/Ajswu5cszpH/Tnpro52nRyCOXbexrYVnMySTvaHOZIYxi/20Fu6vAtGHLCWZM0Y8xLxph1XXxd35sbGWPuNMYsN8YsP3LkyLm3OAm5nA4mzplETUo63rYIJi2NEWMriZgO355zXDH+XNQ0hUgLNuPJzhq0e4rEW3GGj+2lo+2NITrk8cHFW/j2P32Z1hvfCz/7WbeFhyzL4p1Hn+IH91xF1cEdPf9jT2QYGVOYxvYK+wNU708eAiDvy18At5u5F00F4MY3n7bnb3/lK+3Ps9dJ6yJJOzEq5r77YNEi6Lgu6rZtcMstnffF/HnJu8zc9y5Lx9kLdN/11hNE8/IhL69nLyQtTUma9IuA34fH6cCbapf031vfRlW2t30OWyhkL2z90dlFzChNO+35+QE3hxvDOGKdum/vbmDDwSZS3Pbfu7VNYY61RLrshRuOzpqkWZZ1mWVZE7v4+ktvbmRZ1sOWZc20LGtmXk/fWIaRhZOKeKM89ml2fT0Lp4zgUKDD4qE1NYPWlprGEGmhZvWkybBijCGltISGQAZs3hzv5pyTtp8/zLcf/0+CL7wEH/840R/9qMvznl93kCNf+9bJHXPnDlILRYYOp8NQNX08O7OKmbZ5BVGnE8cMe6mOSycU8fVF/8ofrrwD7z+WQH5++/O8Lget7tOHO9aXVtoP/vM/7Z6xe+6BF1+kLSeHrf/yaXjsMViypFMbDtS14Hz+WVzRNt790CdZn2/3ZDjG92Lh+c9+Fr7//V6/fpGupHqdNAUjGKeLHbUtTK0soLKyksLCQlwuF42Nje1l/aOndDDkBTxEohZRC3wuB0t31fPCpmMsGJVJhs9JTXOEY80RsvxK0kDDHRPG3OpcHrrsw/ZGYyPTy7KoyeyQzA7iwtbHGloIhJohI2PQ7imSCEqzUziUkQ+7d8e7Kb229XAj01a+wt68UmZ94je8OHK2/cfZrl0ANAfDLP+3bxAtL2fH937MZdveZscnP0vblq32ECwROc28Ubm8WmH3mjFpcnuBkFSvi/c+fB8L/vQw7rGdC3h015N2uDHU/rghLRP+538IfuVrOGtrGfnyM/aBZcs6XevnS7axcNMbtBUUUnntAn4++yb7QG9G15x3Hlzfq8FPIt3ye1w0hSIcrGsl3GZRmevH7XaTnp5OSUlJe5XI1NRUWlpaaGxspKmpiXA4TH7g5BDGT88fgd/tJNRmceu0fLL9bmqbw9Q2h8kexmujddTXEvw3GGP2AnOAZ4wxf+ufZg0/PreTm2+5iH9880H4wx9wOAxWZcXJEw4dGrS2NNbW4bAsJWky7JRk+dkVyMUagknaSxsOMuXAZrKvuJiV913Ntxd+HKIW/PjHAPztvx9l5g+/jrVnL5/47bdpzsqh8hv34hxZrYICIt24aHQer1VOA8Bx3qxOxyaOyCA3cHqJcI/TQfCUJO1AXQueumPt53z3/PcBYFaeHFodNQ5YurR9u3bLDj57y3ks3PQ6zptu5PzqPDbPv5J9N7wP7r+/f16gSC8FvC6aghFaw3YZfr+nc6+X0+kkNTWVwsJCRo4cSVlZGbm5uViWRcBxcs50ZZaXexeU8tHZhVTnpJCT6qKmKcKxlkiXlSGHo75Wd3zSsqwSy7K8lmUVWJZ1ZX81bDj65wurmP+Vu+1x6UDkW9/hE9ffax8cxJ60hkNH7Qfp6YN2T5FEUJKVwu4hmqTtWr6OnJZ6/PPmEvC6GD1rAovHX4D18MNEGps49PdXAXjw9i8ScbkxP/0p5OSc5aoiw1tptp97H/gEVnV1j3ucHQ5DmzdWkrzVLorw3NqDZLY2Epw0hbZ7PsPrc6+mzTjwhEPUTJjK2tmX8dSkSwktXUZoqd2btva51wnE1k7lrrtI8Th5/rMLGPHE7zREWeLG73XSHGojGFt42ud2dnuuMQafz0dWVhYVFRVMH1fVfizgijKr0MWtE9IIBoNk++2iIsdbNNzxBA13TGBTL5jM2onn2xuDlKRZlkXz4dgC2upJk2GmNNvP/rQ8HHV1UFfXvt968kn4WzcDBUKhTufGi2elvZYTs2cD9tIej42ah6mrY80zr1CwdzsthcXc89tv4ao7TsqtN8extSJDR2VVMWbrVrj66h4/x/LFkrRYT9rqHUdIDzbhvekGnD/8AZ+4dio78+wqztn/8iEif/wjb4+eiaelGc/cOfDCC+xe/a59rZ07YfLkfn1NIucq4HXR2KEnzevqeSpRmJmKM1Y1ZOaEUe1z2bxeLxkeONgQImpBmtv+e3S4U5KWwJwOw8yJZbS4fYOWpB1vDuNuarA3lKTJMFOa5Wd/emwu6J49gP2Lwtx4IyxcCG+9dfqTPvtZe1HZ2tpBbGlndS1hire9S8TjhYl2afALR+eytnAkADWvLmNU7R48k2JVHDus6SQiA+DE4taxJO3Irtjv8Fjv9XtnlFC96GIAzLRpTCvL4mu/+Rqfe8/naUlNo+03v6Vx6w6iDgdmxIjBbr1It/weJ83BNlrDZ+9JO5XTYchP85Luc+HzuPB4PKSnp1NcXMyccWXt5+Wl+Whubm5fSDsYDA7LpE1JWoLLTPVwNJA1aEnavuMtpMUWK1SSJsNNaXYKB9Jz7Y3YkMc31+9pP952yy2d1jh6auUejv3qETh0iMin74FgcFDbe8LGA/WMqD9CqLgE3PaE6/w0Hxmjq6lPzcCzehUja/fhVKl9kUFhdUjSLMuifl/sd3h2h7WjrrzSXkR+ql2YxOP3UXP9e/n7hPnw5ycpPbyHUH4huDT0SxJH6ik9aT5371KJwgwf+em+0/ZPLzu57NOoMns+W0VFBfn5+Xi93vakrbGxkZaWFtra2vr2QoYAJWkJLtXj4pA/E2uQkrT9x1tIC8b+CFWSJsNMms9NsLjE3oglac8//g8AfjXjOpy7dnVaD2nZb/5MVtNx1hZU4/p/j8CMGRA+uVAnr75q7zt2smDAQHj3QD15jbW4Soo77Z8/Oo938quZvOIVfKFWGNeLst0ics4cHZK0g/Wt+BpiQ6I7zgO99VZ77bQO87/njczl0aq5OJubuWLLMtyV5YPYapGzS41Vd2yNnEjSet6TBnD77HI+OOf0/9flOSdHeOQGvBhj7GGQGRkUFxczcuRIKisrKS4uJhAIEA6H25O25uZmQqFQ0vW2KUlLcCkeJ0dTM7EODGKSFlJPmgxfZeOriDicsHMn4bYojavXALDmqpv504xFWD/6Ebz7LseaQlQt/TsRt4en/+t/+fzCT8H69fDkkwBsPtTApn+9F1auhL90saykZcGjj0Jzc9cN+dvfYMOGHrV506EGipprcZeWdNp/04wS1hZUkdUaG8KsJE1kUPg8TkJuL7S0sP1IE1ktsZ/BjkmaMe093ycsnFjImrKJBJ1uPNEIzrIyRBJJqtdFc7CN4Inhjq7eJWk3zSjhA3MqTttvOlQZ7qpqqjEGj8dDIBCgoKCAqqqq9nL/OTk5OJ3O03rbIpHIadcZSpSkJTi/x8lRf+agleDff6yZ3GCjvaEkTYahGVW5rCuoIvz3xWw+1EDZod1EnU4+/fGrefCSO+w5ovfey+vbjjLpwFZap0zjMzfOYOUl17EvuwjrwYcAePS3LzLmnTcAsB5//LT7HH72Rbj9dnj44dOONdbW2XPgJkyAV145a5t3HW2ioKEWU1TUaf/ogjT2X3g5OzOLaLhyEcyceS4hEZFe8rqchNweO0k72kRWS7194CwVVYszU7hlXjWrimNrrylJkwST6nESaovS0GqPGvH2crjjmXx0nr3ge2ZKz9ZJc7lc+P1+srKyKC0t7dTblp6eTjQabV+nzQzBpWaUpCU4v8dJoycF09Q4KPe745M38G9LfgNOp4oLyLA0vTyLF0bNwb38bbas3MTImj20VVRRWZzF7dfO4hfTr8N6+mk2vLKSsUd34Z8+hRSPk89dPYH/O+UqzGuvUvPG22Q/9QRRY3hq3HysF17oVAFy48F6/vDtX9kbzzzT6f6WZfEfX/2/J3d861tnbXPNgSN4w0EoLj7t2L33fYSNr64k7flnThYzEJEB5XM7CLb3pDWSG45NI+g4J60bn14witDcefaGkjRJMKlee45kbZO9OHtve9LO5MtXj2PLt6/C4Ti3hKpjb1teXh4VFRXtc9uKTvkQcyhQkpbgUjwuWtw+TGsrRKMDfr+S3VvsB263FriVYWlCcTovj7XXILKeeoqxtXtwTRwPwM0zS3h81iKiDifTf/F9MlobccRKY186Np/F5y8i5PZy4DvfwcgHbwAAFslJREFUZ/625bROncEjc2/CEQp16jF7c3stF+5cZd/jlVegoaH92O7aZvJWv0WbcfDItEVYS5bA8ePdtjfcFsXat9/e6CJJC3hdLJxY2KeYiEjvZPk9tDjtnrQthxqpNK12AZAerD+a6fcw/2O32hvV1QPcUpHeSfXaSVlNLEnrz540YwxuZ/+mJg6HA6/Xi893erGSRKckLcH53U5a3LGxud3NXRkIsQU4RYYbr8tJYOpEDuSOoHrxX6k6ugcTGyaY6fcwZ95knhs9l8vWLrGfECt573Y6uHzeeJ4cfxEjX/gzUw5uxv+eaym9fD5Lq6YRfeB78PLLAGzcsItJB7eytGwSJhyGl15qv//K3cc4b896msZO4Mnxl2AiEXj++W7be+B4K7n1NfbGEPykUCQZTS3NpNHhprWhkfA7a7jqrefshKunH35edBEsXmxXgBRJIB170ozp3Tpp0juKbILze500u2PZf4fS3wMhFIly3BcY0HuIDAXTK7L5x4iJTN72jr3j/PPbj31gTjm/nt5hUdtYkgbwsYuq+fVlH6TOm4rDsuCqq7hjbgXfn38HLfWNcOmlsGwZ4aXLcFpR/nr1h2hKCXQa8vjOjqNMPbCJwIKLcM2ZzbFAFtZjj3Xb1j3HmslvilWP7KInTUQG37SyTFrdHmqP1nHXsw/jMlZ7UaEeMQYuuQQc+jNNEkuqx07SappCeF2OITnXa6jQT3+C83tctA5ST9qh+lb7D0uAuXMH9F4iiWxGWRZLS+3kyzIGZs1qPzZxRAbmggvYVlRtJ0UdCgFkpXr47EcX8OMv/JjIf9wPM2YwtTSTL3ztDubd+QssY2h65jnyt9pVGwsum8+S8qlEn3nWrvYIHF6+Bn84iOO887h2Rhl/GncRPPWUXaq7C3Vvr+LDy5+yN9STJpIQJhRnEHJ5admznwt2rab2pvepuqokhRM9aTWNQbz9OB9NTqckLcH5PYPXk7bveAsp4Vb23PkpeO21Ab2XSCKbUZ7Fm7EkzYwbd1ql0599YCbpT/wBfve70567YFwB3/zy+3Dd+4X2T8FnVWRTUFXCzhHVNL+4mAmHttFaXskFM6pYXDULx8EDsHgxDa1hfOvskv9Mm8bVk4r409Qr7SGPv/515xtZFlsO1jPh859gysHYXNK0tH6Ng4icG5/biSvgp3rvFrxtEVJvfW+8myTSL/yek3PSeruQtfSOopvgUtxOmk/0pA1wkrb/cB3uaBupOZkqGiLDWk7Ay9S5k6gZM7HLOSE5AS9558+A+fN7fM1rJhexpGAc6SvfZsaBTXhmzWRaaRY18y+l3heAyy5j7ze/y7iD24h6fTB2LNmpHsZcMpullVOJ3n9/e2/asYZW6q55D/5xoynfu5Uls66Ahx7qt9cvIn1XUFoAwKFANpmX9vy9QiSRZcTK4x9vDvd6IWvpHSVpCc7vcQ7acMejB2sBSMvR+mgiP719BjlrV8IDD/TL9W6cXsKKyil4w0EK64/imD4Nh8Pw5Q9fzKKP/Yw3xpzHyAe+wQdWPQuTJ9uV4IBPXzaKry64C6uhAf7t32gORfjDP32GjGefIrvxOOG8fC54+Un45Cf7pZ0i0j9GPPwgex78BUf+/Iy9rI1IEkjvsIZZf5bfl9MpSUtwfo9r0IY71hyxiw+40zVkSgSwl6Lopz+uijNTGH/X+/nz+IvsHRdfDMDI/AA/++w1fOOmL3AwNRtfJIRj0sliJNV5AWZffQE/Pu+98OijvPb+u/nos79g84VXsnzJctxvv4U7VWsaiiSc8nJK7/5nJi44/+znigwRaV5X+2Cr/iy/L6dzxbsBcmY+t4MWz+AkaceOxNZiSk0d0PuIDFd3XjKaxX/8PW0jUnBmnuyxnjgig5/ccwWf8f+SB7Y/T+Wn7+z0vHuvGsd71n+IhZve4Io//ZzjhSWMfur3jM7MHOyXICIiw5jDYUjzuqhvjagnbYApSUtwxhhIiX1KPsDDHZtr6uwHStJEBoTL6eCKCV0vLF2dF+BPX7oGuOa0YwGvi99/+mJ+Mfpxxtbs4j3/dHmPFsUVERHpbxl+N/WtEfWkDTAlaUOAdSJpGuCetEhDo/1ASZpIwskNePnizTOBmfFuioiIDGMZKW720KLCIQNMKfAQYAYpSYs2KkkTERERke6dqPCoJG1gKUkbAlz+FPvBAA53DLdFcbTErq8kTURERES6cCJJ87qURgwkRXcI8PncBD2+Ae1Jq28J4w+32htK0kRERESkCyd70pRGDCRFdwjwe5y0DnCSVtcSxh8O2htK0kRERESkC+m+WJKm6o4DSknaEOD3uGh1+wZ0uGNdS5iUkHrSRERERKR76ZqTNiiUpA0Bfo+TZo93YIc7tkZODnf0a2FcERERETmd5qQNDkV3CPB7nDS7OidplmURikT77R51LWFSwq1EfT5w6pMRERERETmdqjsODq2TNgRU5qbS4PDQWt9Ic1OIddsPwerVfHKLmyUfGk/OutVw3XV9ukf7nDS/hjqKiIiISNdUOGRwKEkbAq6aWMQWtw/fG69RM/M8VuRM4DOv/y//UzKenPs22Cdt2ADjxp3zPepbwhSEWzEBJWkiIiIi0rX24Y7qSRtQSoGHgNJsP970AAAjNr7Dh1c8BUB2cz1Hswrsk7Zv79M96lrCBCLBkwtni4iIiIicojTbT5bfTXVeIN5NSWrqSRsiigPu9seZrY38bPZNPHv7PRzctIO3HroDdu7s0/XrW8KkRYOQriRNRERERLqWneph1VeviHczkp560oaIsk2rAajz2knUO6Om8633TOSIP5OIxwO7dvXp+nZPWkjl90VERERE4kxJ2hDhvOMOAH485xbaMjL54YN3M7kkkwklmRzKyO9zT1pdS5jUSKuSNBERERGROFOSNlQ88ABPvrGVFTd9GLNzJ77MdACunlTMttQ8wtt39OnydS1hMpvqIDe3P1orIiIiIiLnSHPShgpjuGFONTfMqe60e3xxOvvS82Dnqj5d/nhTiIz6Y1BQ0KfriIiIiIhI36gnbYgrTPexLyMfd80RaGk5p2tYlkVL7XHc4SDk5/dzC0VEREREpDeUpA1xhek+9mbEer927oSaml5foynURnpd7HnqSRMRERERias+JWnGmAeMMRuNMWuMMU8aYzL7q2HSM+kpLg7mFtsbv/sd5OXBqt4NfaxpDJLTXGdvqCdNRERERCSu+tqT9iIw0bKsycBm4It9b5L0hjGGlvIqAKK/fQQsC15/vVfXONoYIrfpuL2hnjQRERERkbjqU5JmWdYLlmVFYpvLgJK+N0l6y1eYT503FceunfaOc+hJy2tWkiYiIiIikgj6c07aR4Dn+vF60kNFmSnsyC4+uWP16l49v6YpRM6JnjSV4BcRERERiauzJmnGmJeMMeu6+Lq+wzlfAiLAo2e4zp3GmOXGmOVHjhzpn9YLAAUZPnZmnUzSrHXrIBzu8fNrGoPkNh/HyskBt3sgmigiIiIiIj101nXSLMu67EzHjTEfAq4BFliWZZ3hOg8DDwPMnDmz2/Ok9ywLdmXaSdrK4jFM378JNmyAKVN69PyjjSFGt9ZjVDRERERERCTu+lrdcSHweeA6y7Ka+6dJ0lszy7PYHhvu+PS0K+2dr7zS4+fXNIUobK3TfDQRERERkQTQ1zlpDwFpwIvGmNXGmJ/1Q5ukl66YUMh3fv1l+Na3aHn/B9iRM4K253o+PXDuY79g8o61MHHiALZSRERERER6wpxhhOKAmTlzprV8+fJBv+9w8NqWo2y57SPcsf5FnPv2QXb2WZ9zMLuIuoJixqx6HXy+QWiliIiIiMjwZoxZYVnWzK6O9Wd1R0kAc6pzWD3hfJytrXalxrMMe7QsC3ewhfqKkUrQREREREQSgJK0JON0GHJvuJavLLzbriiybNkZzz/aGMIbCZGSkTZILRQRERERkTNRkpaErptRyiNTFtKSmw8bN57x3B1HGkkJB0nLSh+k1omIiIiIyJkoSUtCk0syqMjxsz2n9KxJ2u6Dx3FaUTKylaSJiIiIiCQCJWlJyBjD9VNHsNJfQPTdjfawx27s2V8LQHq2hjuKiIiIiCQCJWlJ6soJhWzNKcVRdxwOH+72vAMH7CTN4fcPVtNEREREROQMlKQlqbIcP9uyS+yNMwx5PHjouP1ASZqIiIiISEJQkpakAl4XhwvL7I0tW7o9r7mu3n6QkjIIrRIRERERkbNRkpbMSkYQNQ7YvbvbU6JNzfYDJWkiIiIiIglBSVoSK8gOUJOZ222SFm6L4gy22htK0kREREREEoKStCRWkO5jf3pet0lac7ANXyRkb2hOmoiIiIhIQlCSlsQK033sSs3F6iZJawxFSAkH7Q31pImIiIiIJAQlaUmsMMPHvrQ82LMHolEALMvCiq2b1hSMnOxJU5ImIiIiIpIQXPFugAycwnQfG9PzMKEQHD5Mc3YuC+5/iQdrXmdmaQbBS6/DG1FPmoiIiIhIIlGSlsQKM2Jz0gB272bx3iD3//pLzNyxEoDsD23FFzH2cSVpIiIiIiIJQcMdk1h74RCAXbtwff2rXLRjJV+88m5aJkzCvW3ryTlpKhwiIiIiIpIQlKQlsexUD0fSsgEIbdzE5c/+P96afy2Pz1zE9vRCUnbv1Jw0EREREZEEoyQtiTkdBmduLlGHg2PLVuC0oqRffQWXjy/gLWcW/n278YdbsdxucDrj3VwREREREUFJWtLLy0ihIZCJWb8egMrxVbx3egmbUvNxRsJUHNuvXjQRERERkQSiJC3J5aV5OZ6aQdbeHQB4RxRx4ahcjuSXADD2yE4laSIiIiIiCURJWpLLC3g55EnD3Raxd+Tn43I6CFdVA1Bx7ABGRUNERERERBKGkrQkl5/u5XBKxskdeXa1R295CUFnbAUG9aSJiIiIiCQMJWlJLi/gpcafDoCVmQkeDwDF2QG2Z9tDHpWkiYiIiIgkDiVpSS4vzUdtrCfN5Oe37x+RlcKmvHJ7Q0maiIiIiEjCUJKW5PLTvdSkZsY2TiZpxZkpbMqrsDdCocFvmIiIiIiIdElJWpIrSPNRk2IPd+yYpI3ITGHjiSRt27bBb5iIiIiIiHRJSVqSK8vxc+ui6fZGd8Mda2ri0DIREREREemKkrRh4OILJ9oPOiRpuale9qfZlR5PVHwUEREREZH4c8W7ATIIRowAvx/GjGnf5XAYvnfLVHbMe4nKKaPi2DgREREREelISdpwkJ4Ou3ZBdnan3e+dUQIzSuLUKBERERER6YqStOEiNzfeLRARERERkR7QnDQREREREZEEoiRNREREREQkgShJExERERERSSBK0kRERERERBJIn5I0Y8x9xpg1xpjVxpgXjDHF/dUwERERERGR4aivPWkPWJY12bKsqcBfga/2Q5tERERERESGrT4laZZl1XfYTAWsvjVHRERERERkeOvzOmnGmG8DdwB1wCVnOO9O4E6AsrKyvt5WREREREQkKRnLOnPnlzHmJaCwi0NfsizrLx3O+yLgsyzra2e76cyZM63ly5f3tq0iIiIiIiJJwRizwrKsmV0dO2tPmmVZl/XwPo8CzwJnTdJERERERESka30a7miMGWVZ1pbY5vXAxp48b8WKFUeNMbv6cu8BkgscjXcjhiHFPT4U9/hR7ONDcR98inl8KO7xo9jHz1CMfXl3B8463PFMjDGPA2OAKLAL+JhlWfvO+YJxZoxZ3l2XowwcxT0+FPf4UezjQ3EffIp5fCju8aPYx0+yxb5PPWmWZd3UXw0RERERERGRvq+TJiIiIiIiIv1ISVpnD8e7AcOU4h4finv8KPbxobgPPsU8PhT3+FHs4yepYt+nOWkiIiIiIiLSv9STJiIiIiIikkCGdJJmjCk1xrxsjNlgjFlvjPl0bH+2MeZFY8yW2L9Zsf3vN8asMcasNca8YYyZ0uFaC40xm4wxW40x957hnh+MXXeLMeaDHfZ/2xizxxjTOJCvOREkUtw7HH/KGLNuIF5vokiUuBtj0owxqzt8HTXG/NdAv/54ilPsnzfGHDfG/PWU/ZXGmDdjz/+DMcYzUK873vo57r8yxhw+2/tEd98fY8zdsX2WMSZ3oF5zvCVSzDsc/5FJ8t+tiRR3Y8yrHd7f9xtj/jxQrzsRxCn2XZ7X3T2TVX/FvrvrdHPPofMeb1nWkP0CioDpscdpwGZgPPBd4N7Y/nuB/4w9ngtkxR5fBbwZe+wEtgFVgAd4Bxjfxf2yge2xf7Nij09c7/xYexrjHZfhFPfY8RuB/wXWxTs2wynuHc5bAcyPd3ySKfaxcxcA1wJ/PWX/H4H3xR7/DPh4vOOT6HGPbc8Hpp/pfeJM3x9gGlAB7ARy4x2b4RDz2PGZwCMk+e/WRIt7h/MeB+6Id3ySKfZnOq+7eybrV3/FvrvrdHG/IfUeH/cG9PM3+y/A5cAmoKjDN25TF+dmAftij+cAf+tw7IvAF7t4zm3Azzts/xy47ZRzkvoXSaLFHQgAr8V+qJM6SUukuHfYNxrYQ2x+63D5GujYdzh+MR2SNMBgL9Tp6up6yf51rnHvsK/iTO8TPfn+kEC/wJM95th/UL3MMPkANFHi3mFfOnAMSI93PJIp9mc6ryf3TOavvsb+1Ot0sX9IvccP6eGOHRljKrCz4DeBAsuyDsQOHQQKunjKR4HnYo9HYP+hecLe2L5T9fS8YSMB4n4f8H2gufetH7oSIO4nvA/4gxV7ZxsOBin23ckBjluWFTnH5w9ZfYx7T+k9voMEiPndwFMd7jssJEDcT3gP8HfLsup7ee0ha5BifyY9uWdS6q/Yn3KdUw2p9/g+LWadKIwxAewu+Xssy6o3xrQfsyzLMsZYp5x/CfY3d96gNjTJxDvuxpipQLVlWZ+J/VAOC/GO+yneB3xgAK6bkBIs9sOG4j744h1zY0wxcDN2b/KwEe+4n+I24JcDcN2ElGCx7/Keyaq/Yn/qdQa84QNsyPekGWPc2N+QRy3LeiK2+5Axpih2vAg43OH8ydhvOtdbllUT270PKO1w2RJgnzFmdofJs9d1d95AvK5ElyBxnwPMNMbsxB7yONoYs6R/X2liSZC4n7j2FOxhdyv69UUmqEGOfXdqgExjjKvj8/v62hJZP8W9u2uXdoj7x9B7PJAwMZ8GjAS2xt7j/caYrf3yAhNUgsT9xPm5wHnAM31/ZYlvkGN/Jt3eM1n1V+y7us6Qf4+P93jLvnxhz8/4LfBfp+x/gM4TDr8be1wGbAXmnnK+C7soQiUnJxJO6OJ+2cAO7HGwWbHH2aeck/Tj5hM07hUk+Zy0RIs7cD/wjXjHJRlj3+H8izm9cMhjdC4c8ol4xyfR497heWd8n+jJ94cEmq8wXGIeOy+pf7cmWtyBjwG/iXdckjH2Zzqvu3sm61d/xb6763RxvyH1Hh/3BvTxmzsPsIA1wOrY1yLseRt/B7YALxH7wxI78z7W4dzlHa61CLsazDbgS2e450di/0G2Ah/usP+72GNbo7F/vx7v+AyHuHc43qM3xaH8lWhxj73RjY13XJI49q8CR4CW2HvKlbH9VcBbse/JY4A33vEZInH/HXAACMfi+dFu7tnl9wf4VOx5EWA/8Mt4xyfZY37KOcmepCVU3IElwMJ4xyWJY9/led3dM1m/+iv23V2nm3sOmfd4E2uYiIiIiIiIJIAhPydNREREREQkmShJExERERERSSBK0kRERERERBKIkjQREREREZEEoiRNREREREQkgShJExERERERSSBK0kRERERERBKIkjQREREREZEE8v8B+iNtULSqtHkAAAAASUVORK5CYII=",
            "text/plain": [
              "<Figure size 1080x360 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          },
          "output_type": "display_data"
        }
      ],
      "source": [
        "## First, let's see how this does with the AIC selected values. \n",
        "arma = ARIMA(train_price, order=min_aic_index).fit()\n",
        "print(arma.summary())\n",
        "fig, ax = plt.subplots(figsize=(15, 5))\n",
        "\n",
        "# Construct the forecasts\n",
        "fcast = arma.get_forecast(len(test_price)).summary_frame()\n",
        "\n",
        "arma_predictions = arma.predict()\n",
        "ax.plot(california.index.values, cubic_residuals, label='log cases, converted to stationary time series')\n",
        "predicted_values = arma_predictions\n",
        "ax.plot(train_date, predicted_values, 'r', label='fitted line')\n",
        "forecast_means = fcast['mean'].values.reshape(-1,1)\n",
        "test_set_mse = np.mean((forecast_means.reshape(test_price.shape) - test_price)**2)\n",
        "ax.plot(test_date, forecast_means, 'k--', label='mean forecast')\n",
        "ax.fill_between(test_date.flatten(), fcast['mean_ci_lower'], fcast['mean_ci_upper'], color='k', alpha=0.1);\n",
        "plt.legend()\n",
        "print(\"Test set mean squared error: \", test_set_mse)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "8fACkKgVlvMz"
      },
      "source": [
        "In previous ACF and PACF plots, it seems that the AR order is 1. Let's try to use AR(1) and MA(3) model and compare the two models by MSE."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 28,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 931
        },
        "id": "2dIZRBholhhJ",
        "outputId": "65f87ac2-322e-4e85-a170-6753821f9c8a"
      },
      "outputs": [
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/statsmodels/tsa/base/tsa_model.py:471: ValueWarning: No frequency information was provided, so inferred frequency D will be used.\n",
            "  self._init_dates(dates, freq)\n",
            "/usr/local/lib/python3.7/dist-packages/statsmodels/tsa/base/tsa_model.py:471: ValueWarning: No frequency information was provided, so inferred frequency D will be used.\n",
            "  self._init_dates(dates, freq)\n",
            "/usr/local/lib/python3.7/dist-packages/statsmodels/tsa/base/tsa_model.py:471: ValueWarning: No frequency information was provided, so inferred frequency D will be used.\n",
            "  self._init_dates(dates, freq)\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "                               SARIMAX Results                                \n",
            "==============================================================================\n",
            "Dep. Variable:              JHU_cases   No. Observations:                  591\n",
            "Model:                 ARIMA(1, 0, 3)   Log Likelihood                 491.103\n",
            "Date:                Sun, 28 Aug 2022   AIC                           -970.206\n",
            "Time:                        00:32:07   BIC                           -943.915\n",
            "Sample:                    01-01-2020   HQIC                          -959.964\n",
            "                         - 08-13-2021                                         \n",
            "Covariance Type:                  opg                                         \n",
            "==============================================================================\n",
            "                 coef    std err          z      P>|z|      [0.025      0.975]\n",
            "------------------------------------------------------------------------------\n",
            "const          0.3036      0.586      0.518      0.605      -0.846       1.453\n",
            "ar.L1          0.9887      0.006    162.221      0.000       0.977       1.001\n",
            "ma.L1          0.2304      0.037      6.297      0.000       0.159       0.302\n",
            "ma.L2          0.1903      0.036      5.260      0.000       0.119       0.261\n",
            "ma.L3          0.1282      0.036      3.513      0.000       0.057       0.200\n",
            "sigma2         0.0110      0.000     35.788      0.000       0.010       0.012\n",
            "===================================================================================\n",
            "Ljung-Box (L1) (Q):                   0.22   Jarque-Bera (JB):              1706.77\n",
            "Prob(Q):                              0.64   Prob(JB):                         0.00\n",
            "Heteroskedasticity (H):               1.15   Skew:                             0.39\n",
            "Prob(H) (two-sided):                  0.33   Kurtosis:                        11.29\n",
            "===================================================================================\n",
            "\n",
            "Warnings:\n",
            "[1] Covariance matrix calculated using the outer product of gradients (complex-step).\n",
            "Test set mean squared error:  0.9221242342161149\n"
          ]
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 1080x360 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          },
          "output_type": "display_data"
        }
      ],
      "source": [
        "## Then, let's see how this does with the AR(1) and MA(3). \n",
        "arma = ARIMA(train_price, order=(1,0,3)).fit()\n",
        "print(arma.summary())\n",
        "fig, ax = plt.subplots(figsize=(15, 5))\n",
        "\n",
        "# Construct the forecasts\n",
        "fcast = arma.get_forecast(len(test_price)).summary_frame()\n",
        "\n",
        "arma_predictions = arma.predict()\n",
        "ax.plot(california.index.values, cubic_residuals, label='log cases, converted to stationary time series')\n",
        "predicted_values = arma_predictions\n",
        "ax.plot(train_date, predicted_values, 'r', label='fitted line')\n",
        "forecast_means = fcast['mean'].values.reshape(-1,1)\n",
        "test_set_mse = np.mean((forecast_means.reshape(test_price.shape) - test_price)**2)\n",
        "ax.plot(test_date, forecast_means, 'k--', label='mean forecast')\n",
        "ax.fill_between(test_date.flatten(), fcast['mean_ci_lower'], fcast['mean_ci_upper'], color='k', alpha=0.1);\n",
        "plt.legend()\n",
        "print(\"Test set mean squared error: \", test_set_mse)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "fXfzqPCjmNnN"
      },
      "source": [
        "The best predictor of the cases uses the AR(1) model and has coefficient of 0.9887. That is, the best predictor of COVID-19 cases is 0.9887 times today's COVID-19 acses (after taking the rolling average, log transformation, and removing the cubic trend). It turns out that ACF/PACF is more accurate than just doing a grid search over values."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 29,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "5L-6iefh2J0U",
        "outputId": "dd5c961e-03a3-433c-cddc-bb034b82e0b6"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "MSE of ARIMA model for weekly prediction: 0.1949\n"
          ]
        }
      ],
      "source": [
        "# predicted_values - week_cubic_residuals\n",
        "# forecast_means - week_test\n",
        "\n",
        "predicted_labels = np.concatenate((predicted_values.reset_index().predicted_mean, forecast_means.flatten()), axis=None)\n",
        "true_labels = cubic_residuals.reset_index().JHU_cases\n",
        "\n",
        "week_arima_mse = mean_squared_error(true_labels, predicted_labels)\n",
        "print(f\"MSE of ARIMA model for weekly prediction: {week_arima_mse:.4f}\")"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "gfHXOU6zm6qV"
      },
      "source": [
        "## 9. Weekly Forecasting\n",
        "\n",
        "### Extension: Weekly Predictions\n",
        "\n",
        "If you're interested in long-range forecasts, it might be better to convert the time series to weekly values. \n",
        "\n",
        "This loses lots of rich variation, but avoids the problem where predictions one week away will have converged to the mean of the series. \n",
        "\n",
        "We follow the same steps: convert to stationary, model the AR/MA components, and then form predictions. "
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 30,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 390
        },
        "id": "Lwzs0iChntf7",
        "outputId": "b2aee090-d7d4-451c-9c5c-c18df9517746"
      },
      "outputs": [
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 864x432 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          },
          "output_type": "display_data"
        }
      ],
      "source": [
        "collapsed = log_cases.groupby(pd.Grouper(freq ='W')).mean()\n",
        "week_date = collapsed.reset_index().date.dt.date.values.reshape(-1,1)\n",
        "week_log_case = collapsed.reset_index().JHU_cases.values.reshape(-1,1)\n",
        "fig, ax = plt.subplots(1,1, figsize = (12,6))\n",
        "plt.plot(week_date, week_log_case)\n",
        "plt.title(\"log of COVID-19 cases over time\")\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 31,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 751
        },
        "id": "J4wDpazwqJpw",
        "outputId": "bd98f122-3c37-4a97-de49-917993376ad5"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "[[ 6.03914116e-01 -1.07353067e-02  5.75986373e-05]]\n"
          ]
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 864x864 with 2 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          },
          "output_type": "display_data"
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "MSE with cubic fit: 0.9587963703481652\n"
          ]
        }
      ],
      "source": [
        "clf = linear_model.LinearRegression()\n",
        "index = collapsed.reset_index().index.values.reshape(-1,1)\n",
        "new_x = np.hstack((index, index **2, index **3))\n",
        "clf.fit(new_x, week_log_case)\n",
        "print(clf.coef_) # To print the coefficient estimate of the series. \n",
        "month_quad_prediction = clf.predict(new_x)\n",
        "\n",
        "fig, axs = plt.subplots(2,1, figsize=(12,12))\n",
        "axs[0].plot(week_date, week_log_case, label='original data')\n",
        "axs[0].plot(week_date, month_quad_prediction, 'r', label='fitted line')\n",
        "axs[0].legend()\n",
        "axs[0].set_title(\"True Log Cases vs Fitting Cubic Line\")\n",
        "\n",
        "week_cubic_residuals = week_log_case - month_quad_prediction\n",
        "axs[1].plot(week_date, week_cubic_residuals, 'o')\n",
        "axs[0].set_title(\"Residuals: True Log Cases - Fitting Cubic Line\")\n",
        "plt.show();\n",
        "print(\"MSE with cubic fit:\", np.mean((week_cubic_residuals)**2))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 32,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 771
        },
        "id": "XeGCcfERq6S0",
        "outputId": "2da610b6-76ab-4e48-b3a9-75c8c0d459b7"
      },
      "outputs": [
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/statsmodels/graphics/tsaplots.py:353: FutureWarning: The default method 'yw' can produce PACF values outside of the [-1,1] interval. After 0.13, the default will change tounadjusted Yule-Walker ('ywm'). You can use this method now by setting method='ywm'.\n",
            "  FutureWarning,\n"
          ]
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 864x864 with 2 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          },
          "output_type": "display_data"
        }
      ],
      "source": [
        "fig, axs = plt.subplots(2,1, figsize = (12,12))\n",
        "sm.graphics.tsa.plot_acf(week_cubic_residuals, lags=14, ax = axs[0])\n",
        "sm.graphics.tsa.plot_pacf(week_cubic_residuals, lags=14, ax = axs[1])\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 33,
      "metadata": {
        "id": "JK5cYWt_rpZe"
      },
      "outputs": [],
      "source": [
        "week_train_test = int(0.8 * len(week_date))\n",
        "week_train, week_test = week_cubic_residuals[:week_train_test], week_cubic_residuals[week_train_test:]\n",
        "week_date_train, week_date_test = week_date[:week_train_test], week_date[week_train_test:]"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 34,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "TryfM_d1rSZ8",
        "outputId": "dfc973c5-f819-41fb-93c4-9d883e0a1e70"
      },
      "outputs": [
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/statsmodels/tsa/statespace/sarimax.py:978: UserWarning: Non-invertible starting MA parameters found. Using zeros as starting parameters.\n",
            "  warn('Non-invertible starting MA parameters found.'\n",
            "/usr/local/lib/python3.7/dist-packages/statsmodels/base/model.py:606: ConvergenceWarning: Maximum Likelihood optimization failed to converge. Check mle_retvals\n",
            "  ConvergenceWarning)\n",
            "/usr/local/lib/python3.7/dist-packages/statsmodels/base/model.py:606: ConvergenceWarning: Maximum Likelihood optimization failed to converge. Check mle_retvals\n",
            "  ConvergenceWarning)\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Minimizing AIC order:  (2, 0, 1)\n",
            "Minimizing BIC order:  (2, 0, 1)\n",
            "matrix of AIC [[300.35485347 185.65100517 116.01760485  80.42558629]\n",
            " [ 84.226699    51.18087919  40.16808387  35.84386527]\n",
            " [ 24.44400441  19.25152136  21.24784073  23.24307942]\n",
            " [ 20.43466682  21.24797197  19.28397092  21.44216753]]\n",
            "Matrix of BIC [[305.68173166 193.64132245 126.67136123  93.74278176]\n",
            " [ 92.21701628  61.83463557  53.48527934  51.82449984]\n",
            " [ 35.09776079  32.56871683  37.2284753   41.88715308]\n",
            " [ 33.75186229  37.22860653  37.92804458  42.74968028]]\n",
            "                               SARIMAX Results                                \n",
            "==============================================================================\n",
            "Dep. Variable:                      y   No. Observations:                  106\n",
            "Model:                 ARIMA(2, 0, 1)   Log Likelihood                  -4.626\n",
            "Date:                Sun, 28 Aug 2022   AIC                             19.252\n",
            "Time:                        00:32:12   BIC                             32.569\n",
            "Sample:                             0   HQIC                            24.649\n",
            "                                - 106                                         \n",
            "Covariance Type:                  opg                                         \n",
            "==============================================================================\n",
            "                 coef    std err          z      P>|z|      [0.025      0.975]\n",
            "------------------------------------------------------------------------------\n",
            "const          0.0993      0.200      0.495      0.620      -0.293       0.492\n",
            "ar.L1          1.7762      0.062     28.691      0.000       1.655       1.898\n",
            "ar.L2         -0.8654      0.064    -13.526      0.000      -0.991      -0.740\n",
            "ma.L1         -0.3877      0.144     -2.686      0.007      -0.670      -0.105\n",
            "sigma2         0.0617      0.008      7.854      0.000       0.046       0.077\n",
            "===================================================================================\n",
            "Ljung-Box (L1) (Q):                   0.05   Jarque-Bera (JB):                66.30\n",
            "Prob(Q):                              0.82   Prob(JB):                         0.00\n",
            "Heteroskedasticity (H):               0.59   Skew:                             0.35\n",
            "Prob(H) (two-sided):                  0.12   Kurtosis:                         6.81\n",
            "===================================================================================\n",
            "\n",
            "Warnings:\n",
            "[1] Covariance matrix calculated using the outer product of gradients (complex-step).\n"
          ]
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 864x864 with 2 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          },
          "output_type": "display_data"
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Automatic selection finds model with AR 2, MA 1\n",
            "MSE with selected model: 0.07337165944319171\n"
          ]
        }
      ],
      "source": [
        "min_aic_index, min_bic_index, *other = grid_search_ARIMA(week_cubic_residuals, range(4), range(4), verbose=True)\n",
        "if min_aic_index == min_bic_index:\n",
        "  arma = ARIMA(week_cubic_residuals, order=min_bic_index).fit()\n",
        "  print(arma.summary())\n",
        "  arma_predictions = arma.predict()\n",
        "  arma_residuals = week_cubic_residuals.reshape(arma_predictions.shape) - arma_predictions\n",
        "  arma_residuals = arma_residuals # Fitting AR 1 model means removing one observation\n",
        "  \n",
        "  fig, axs = plt.subplots(2,1, figsize = (12,12))\n",
        "  axs[0].plot(week_cubic_residuals, label='Residuals from fitted cubic line')\n",
        "  axs[0].plot(arma_predictions, 'r', label='fitted ARMA process')\n",
        "  axs[0].legend()\n",
        "  axs[1].plot(arma_residuals, 'o')\n",
        "  plt.show()\n",
        "  print(\"Automatic selection finds model with AR {0}, MA {2}\".format(*min_aic_index))\n",
        "  print(\"MSE with selected model:\", np.mean(arma_residuals**2))\n",
        "else:\n",
        "  print(\"AIC, BIC do not agree.\")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 35,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 337
        },
        "id": "jIoFrfY_ren4",
        "outputId": "df962906-3118-4650-803d-3fa4596186b2"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Test set mean squared error:  0.4762984669587506\n"
          ]
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 1080x360 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          },
          "output_type": "display_data"
        }
      ],
      "source": [
        "arma = ARIMA(week_train, order=min_bic_index).fit()\n",
        "fig, ax = plt.subplots(figsize=(15, 5))\n",
        "\n",
        "# Construct the forecasts\n",
        "fcast = arma.get_forecast(len(week_test)).summary_frame()\n",
        "\n",
        "arma_predictions = arma.predict()\n",
        "ax.plot(week_date, week_cubic_residuals, label='original data')\n",
        "predicted_values = arma_predictions.reshape(-1,1)\n",
        "ax.plot(week_date_train, predicted_values, 'r', label='fitted line')\n",
        "forecast_means = fcast['mean'].values.reshape(-1,1)\n",
        "ax.plot(week_date_test, forecast_means, 'k--', label='mean forecast')\n",
        "ax.fill_between(week_date_test.flatten(), fcast['mean_ci_lower'], fcast['mean_ci_upper'], color='k', alpha=0.1);\n",
        "plt.legend();\n",
        "\n",
        "test_set_mse = np.mean((forecast_means.reshape(week_test.shape) - week_test)**2)\n",
        "print(\"Test set mean squared error: \", test_set_mse)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 36,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "cWzf25aayqMG",
        "outputId": "a9ce9def-09ca-40f4-da87-258346132c64"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "MSE of ARIMA model for weekly prediction: 0.1621\n"
          ]
        }
      ],
      "source": [
        "# predicted_values - week_cubic_residuals\n",
        "# forecast_means - week_test\n",
        "predicted_labels = np.concatenate((predicted_values.flatten(), forecast_means.flatten()), axis=None)\n",
        "true_labels = week_cubic_residuals.flatten()\n",
        "\n",
        "week_arima_mse = mean_squared_error(true_labels, predicted_labels)\n",
        "print(f\"MSE of ARIMA model for weekly prediction: {week_arima_mse:.4f}\")\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "## Reference \n",
        "\n",
        "Benvenuto, Domenico, et al. \"Application of the ARIMA model on the COVID-2019 epidemic dataset.\" Data in brief 29 (2020): 105340."
      ]
    }
  ],
  "metadata": {
    "colab": {
      "collapsed_sections": [],
      "name": "time-series-analysis.ipynb",
      "provenance": []
    },
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}