{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "%matplotlib inline\n",
    "%config InlineBackend.figure_format = 'retina'\n",
    "\n",
    "import numpy as np\n",
    "import matplotlib\n",
    "from matplotlib import pyplot as plt\n",
    "import pandas as pd\n",
    "from datetime import date, datetime\n",
    "from lifelines import KaplanMeierFitter, CoxPHFitter, NelsonAalenFitter\n",
    "\n",
    "matplotlib.rcParams['figure.figsize'] = (12.0, 6.0)\n",
    "plt.style.use('seaborn-deep')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Definition of censoring and death\n",
    "\n",
    "Quitting is death, all else is censoring. This is different than the [original article](https://fivethirtyeight.com/features/two-years-in-turnover-in-trumps-cabinet-is-still-historically-high/)'s author's rules, who stated that switching roles _within_ a cabinent is an \"event\". "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<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>president</th>\n",
       "      <th>president_start_date</th>\n",
       "      <th>president_end_date</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Trump</td>\n",
       "      <td>2017-01-20</td>\n",
       "      <td>2020-07-26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Obama</td>\n",
       "      <td>2009-01-20</td>\n",
       "      <td>2017-01-20</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Bush 43</td>\n",
       "      <td>2001-01-20</td>\n",
       "      <td>2009-01-20</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Clinton</td>\n",
       "      <td>1993-01-20</td>\n",
       "      <td>2001-01-20</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Bush 41</td>\n",
       "      <td>1989-01-20</td>\n",
       "      <td>1993-01-20</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>Reagan</td>\n",
       "      <td>1981-01-20</td>\n",
       "      <td>1989-01-20</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>Carter</td>\n",
       "      <td>1977-01-20</td>\n",
       "      <td>1981-01-20</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  president president_start_date president_end_date\n",
       "0     Trump           2017-01-20         2020-07-26\n",
       "1     Obama           2009-01-20         2017-01-20\n",
       "2   Bush 43           2001-01-20         2009-01-20\n",
       "3   Clinton           1993-01-20         2001-01-20\n",
       "4   Bush 41           1989-01-20         1993-01-20\n",
       "5    Reagan           1981-01-20         1989-01-20\n",
       "6    Carter           1977-01-20         1981-01-20"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "raw_df = pd.read_csv(\"https://raw.githubusercontent.com/fivethirtyeight/data/master/cabinet-turnover/cabinet-turnover.csv\",\n",
    "                    na_values=['Still in office', '#VALUE!']\n",
    "                    )\n",
    "TODAY = datetime.today().date()\n",
    "\n",
    "INAUG_DATES = {\n",
    "    'Trump': date(2017, 1, 20),\n",
    "    'Obama': date(2009, 1, 20),\n",
    "    'Bush 43': date(2001, 1, 20),\n",
    "    'Clinton': date(1993, 1, 20),\n",
    "    'Bush 41': date(1989, 1, 20),\n",
    "    'Reagan': date(1981, 1, 20),\n",
    "    'Carter': date(1977, 1, 20)\n",
    "}\n",
    "\n",
    "presidential_terms = pd.DataFrame(list(INAUG_DATES.items()))\n",
    "presidential_terms.columns = ['president', 'president_start_date']\n",
    "presidential_terms['president_end_date'] = presidential_terms['president_start_date'].shift(1).fillna(TODAY)\n",
    "presidential_terms"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "def fill_end(series):\n",
    "    end, president = series\n",
    "    if pd.notnull(end) and end.endswith('admin'):\n",
    "        next_pres ,_ = end.split(' ')\n",
    "        if next_pres == 'Bush':\n",
    "            next_pres = next_pres + ' 43' if president == 'Clinton' else next_pres + ' 41'\n",
    "        return INAUG_DATES[next_pres].strftime('%m/%d/%y')\n",
    "    else:\n",
    "        return end\n",
    "    \n",
    "def fill_start(series):\n",
    "    end, president = series\n",
    "    if pd.notnull(end) and end.endswith('admin'):\n",
    "        prev_pres ,_ = end.split(' ')\n",
    "        if prev_pres == 'Bush':\n",
    "            prev_pres = prev_pres + ' 43' if president == 'Obama' else prev_pres + ' 41'\n",
    "        return INAUG_DATES[president].strftime('%m/%d/%y')\n",
    "    else:\n",
    "        return end\n",
    "    \n",
    "    \n",
    "raw_df['end'] = raw_df[['end', 'president']].apply(fill_end, axis=1)\n",
    "raw_df['start'] = raw_df[['start', 'president']].apply(fill_start, axis=1)\n",
    "\n",
    "raw_df['end'] = pd.to_datetime(raw_df['end']).dt.date\n",
    "raw_df['end'] = raw_df['end'].fillna(TODAY)\n",
    "raw_df['start'] = pd.to_datetime(raw_df['start']).dt.date"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "raw_df = raw_df.merge(presidential_terms, left_on='president', right_on='president')\n",
    "raw_df['event'] = (raw_df['end'] < raw_df['president_end_date']) & pd.notnull(raw_df['end'])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "# we need to \"collapse\" individuals into rows, because they may change positions, but that's not quitting...\n",
    "def collapse(df):\n",
    "    return df.groupby('appointee', as_index=False).aggregate({\n",
    "        'start': 'min', 'end': 'max', 'event': 'all', 'president': 'last', 'president_end_date': 'last'\n",
    "    })\n",
    "\n",
    "raw_df = raw_df.groupby('president', as_index=False).apply(collapse).reset_index(drop=True)\n",
    "raw_df['T'] = (raw_df['end'] - raw_df['start']).dt.days\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<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>appointee</th>\n",
       "      <th>start</th>\n",
       "      <th>end</th>\n",
       "      <th>event</th>\n",
       "      <th>president</th>\n",
       "      <th>president_end_date</th>\n",
       "      <th>T</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>267</th>\n",
       "      <td>Jeff Sessions</td>\n",
       "      <td>2017-02-09</td>\n",
       "      <td>2018-11-07</td>\n",
       "      <td>True</td>\n",
       "      <td>Trump</td>\n",
       "      <td>2020-07-26</td>\n",
       "      <td>636</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>268</th>\n",
       "      <td>Jim Mattis</td>\n",
       "      <td>2017-01-20</td>\n",
       "      <td>2018-12-31</td>\n",
       "      <td>True</td>\n",
       "      <td>Trump</td>\n",
       "      <td>2020-07-26</td>\n",
       "      <td>710</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>269</th>\n",
       "      <td>John Kelly</td>\n",
       "      <td>2017-01-20</td>\n",
       "      <td>2018-12-31</td>\n",
       "      <td>True</td>\n",
       "      <td>Trump</td>\n",
       "      <td>2020-07-26</td>\n",
       "      <td>710</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>270</th>\n",
       "      <td>Kirstjen Nielsen</td>\n",
       "      <td>2017-12-06</td>\n",
       "      <td>2020-07-26</td>\n",
       "      <td>False</td>\n",
       "      <td>Trump</td>\n",
       "      <td>2020-07-26</td>\n",
       "      <td>963</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>271</th>\n",
       "      <td>Linda McMahon</td>\n",
       "      <td>2017-02-14</td>\n",
       "      <td>2020-07-26</td>\n",
       "      <td>False</td>\n",
       "      <td>Trump</td>\n",
       "      <td>2020-07-26</td>\n",
       "      <td>1258</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>272</th>\n",
       "      <td>Mick Mulvaney</td>\n",
       "      <td>2017-02-16</td>\n",
       "      <td>2020-07-26</td>\n",
       "      <td>False</td>\n",
       "      <td>Trump</td>\n",
       "      <td>2020-07-26</td>\n",
       "      <td>1256</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>273</th>\n",
       "      <td>Mike Pence</td>\n",
       "      <td>2017-01-20</td>\n",
       "      <td>2020-07-26</td>\n",
       "      <td>False</td>\n",
       "      <td>Trump</td>\n",
       "      <td>2020-07-26</td>\n",
       "      <td>1283</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>274</th>\n",
       "      <td>Mike Pompeo</td>\n",
       "      <td>2017-01-23</td>\n",
       "      <td>2020-07-26</td>\n",
       "      <td>False</td>\n",
       "      <td>Trump</td>\n",
       "      <td>2020-07-26</td>\n",
       "      <td>1280</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>275</th>\n",
       "      <td>Nikki Haley</td>\n",
       "      <td>2017-01-27</td>\n",
       "      <td>2018-12-31</td>\n",
       "      <td>True</td>\n",
       "      <td>Trump</td>\n",
       "      <td>2020-07-26</td>\n",
       "      <td>703</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>276</th>\n",
       "      <td>Reince Priebus</td>\n",
       "      <td>2017-01-20</td>\n",
       "      <td>2017-07-28</td>\n",
       "      <td>True</td>\n",
       "      <td>Trump</td>\n",
       "      <td>2020-07-26</td>\n",
       "      <td>189</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>277</th>\n",
       "      <td>Rex Tillerson</td>\n",
       "      <td>2017-02-01</td>\n",
       "      <td>2018-03-31</td>\n",
       "      <td>True</td>\n",
       "      <td>Trump</td>\n",
       "      <td>2020-07-26</td>\n",
       "      <td>423</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>278</th>\n",
       "      <td>Rick Perry</td>\n",
       "      <td>2017-03-02</td>\n",
       "      <td>2020-07-26</td>\n",
       "      <td>False</td>\n",
       "      <td>Trump</td>\n",
       "      <td>2020-07-26</td>\n",
       "      <td>1242</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>279</th>\n",
       "      <td>Robert Lighthizer</td>\n",
       "      <td>2017-05-15</td>\n",
       "      <td>2020-07-26</td>\n",
       "      <td>False</td>\n",
       "      <td>Trump</td>\n",
       "      <td>2020-07-26</td>\n",
       "      <td>1168</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>280</th>\n",
       "      <td>Robert Wilkie</td>\n",
       "      <td>2018-07-30</td>\n",
       "      <td>2020-07-26</td>\n",
       "      <td>False</td>\n",
       "      <td>Trump</td>\n",
       "      <td>2020-07-26</td>\n",
       "      <td>727</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>281</th>\n",
       "      <td>Ryan Zinke</td>\n",
       "      <td>2017-03-01</td>\n",
       "      <td>2019-01-02</td>\n",
       "      <td>True</td>\n",
       "      <td>Trump</td>\n",
       "      <td>2020-07-26</td>\n",
       "      <td>672</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>282</th>\n",
       "      <td>Scott Pruitt</td>\n",
       "      <td>2017-02-17</td>\n",
       "      <td>2018-07-06</td>\n",
       "      <td>True</td>\n",
       "      <td>Trump</td>\n",
       "      <td>2020-07-26</td>\n",
       "      <td>504</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>283</th>\n",
       "      <td>Sonny Perdue</td>\n",
       "      <td>2017-04-25</td>\n",
       "      <td>2020-07-26</td>\n",
       "      <td>False</td>\n",
       "      <td>Trump</td>\n",
       "      <td>2020-07-26</td>\n",
       "      <td>1188</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>284</th>\n",
       "      <td>Steve Mnuchin</td>\n",
       "      <td>2017-02-13</td>\n",
       "      <td>2020-07-26</td>\n",
       "      <td>False</td>\n",
       "      <td>Trump</td>\n",
       "      <td>2020-07-26</td>\n",
       "      <td>1259</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>285</th>\n",
       "      <td>Tom Price</td>\n",
       "      <td>2017-02-10</td>\n",
       "      <td>2017-09-29</td>\n",
       "      <td>True</td>\n",
       "      <td>Trump</td>\n",
       "      <td>2020-07-26</td>\n",
       "      <td>231</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>286</th>\n",
       "      <td>Wilbur Ross</td>\n",
       "      <td>2017-02-28</td>\n",
       "      <td>2020-07-26</td>\n",
       "      <td>False</td>\n",
       "      <td>Trump</td>\n",
       "      <td>2020-07-26</td>\n",
       "      <td>1244</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             appointee       start         end  event president  \\\n",
       "267      Jeff Sessions  2017-02-09  2018-11-07   True     Trump   \n",
       "268         Jim Mattis  2017-01-20  2018-12-31   True     Trump   \n",
       "269         John Kelly  2017-01-20  2018-12-31   True     Trump   \n",
       "270   Kirstjen Nielsen  2017-12-06  2020-07-26  False     Trump   \n",
       "271      Linda McMahon  2017-02-14  2020-07-26  False     Trump   \n",
       "272      Mick Mulvaney  2017-02-16  2020-07-26  False     Trump   \n",
       "273         Mike Pence  2017-01-20  2020-07-26  False     Trump   \n",
       "274        Mike Pompeo  2017-01-23  2020-07-26  False     Trump   \n",
       "275        Nikki Haley  2017-01-27  2018-12-31   True     Trump   \n",
       "276     Reince Priebus  2017-01-20  2017-07-28   True     Trump   \n",
       "277      Rex Tillerson  2017-02-01  2018-03-31   True     Trump   \n",
       "278         Rick Perry  2017-03-02  2020-07-26  False     Trump   \n",
       "279  Robert Lighthizer  2017-05-15  2020-07-26  False     Trump   \n",
       "280      Robert Wilkie  2018-07-30  2020-07-26  False     Trump   \n",
       "281         Ryan Zinke  2017-03-01  2019-01-02   True     Trump   \n",
       "282       Scott Pruitt  2017-02-17  2018-07-06   True     Trump   \n",
       "283       Sonny Perdue  2017-04-25  2020-07-26  False     Trump   \n",
       "284      Steve Mnuchin  2017-02-13  2020-07-26  False     Trump   \n",
       "285          Tom Price  2017-02-10  2017-09-29   True     Trump   \n",
       "286        Wilbur Ross  2017-02-28  2020-07-26  False     Trump   \n",
       "\n",
       "    president_end_date     T  \n",
       "267         2020-07-26   636  \n",
       "268         2020-07-26   710  \n",
       "269         2020-07-26   710  \n",
       "270         2020-07-26   963  \n",
       "271         2020-07-26  1258  \n",
       "272         2020-07-26  1256  \n",
       "273         2020-07-26  1283  \n",
       "274         2020-07-26  1280  \n",
       "275         2020-07-26   703  \n",
       "276         2020-07-26   189  \n",
       "277         2020-07-26   423  \n",
       "278         2020-07-26  1242  \n",
       "279         2020-07-26  1168  \n",
       "280         2020-07-26   727  \n",
       "281         2020-07-26   672  \n",
       "282         2020-07-26   504  \n",
       "283         2020-07-26  1188  \n",
       "284         2020-07-26  1259  \n",
       "285         2020-07-26   231  \n",
       "286         2020-07-26  1244  "
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "raw_df.tail(20)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<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",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>model</th>\n",
       "      <td>lifelines.PiecewiseExponentialFitter</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>number of observations</th>\n",
       "      <td>287</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>number of events observed</th>\n",
       "      <td>158</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>log-likelihood</th>\n",
       "      <td>-1313.4569</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>hypothesis</th>\n",
       "      <td>lambda_0_ != 1, lambda_1_ != 1, lambda_2_ != 1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div><table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>coef</th>\n",
       "      <th>se(coef)</th>\n",
       "      <th>coef lower 95%</th>\n",
       "      <th>coef upper 95%</th>\n",
       "      <th>z</th>\n",
       "      <th>p</th>\n",
       "      <th>-log2(p)</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>lambda_0_</th>\n",
       "      <td>3067.6037</td>\n",
       "      <td>310.1869</td>\n",
       "      <td>2459.6485</td>\n",
       "      <td>3675.5588</td>\n",
       "      <td>9.8863</td>\n",
       "      <td>&lt;5e-05</td>\n",
       "      <td>74.1495</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>lambda_1_</th>\n",
       "      <td>153.3781</td>\n",
       "      <td>29.9346</td>\n",
       "      <td>94.7075</td>\n",
       "      <td>212.0488</td>\n",
       "      <td>5.0904</td>\n",
       "      <td>&lt;5e-05</td>\n",
       "      <td>21.4161</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>lambda_2_</th>\n",
       "      <td>1038.9781</td>\n",
       "      <td>168.4720</td>\n",
       "      <td>708.7791</td>\n",
       "      <td>1369.1771</td>\n",
       "      <td>6.1611</td>\n",
       "      <td>&lt;5e-05</td>\n",
       "      <td>30.3667</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table><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",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>AIC</th>\n",
       "      <td>2632.9137</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/latex": [
       "\\begin{tabular}{lrrrrrrr}\n",
       "\\toprule\n",
       "{} &      coef &  se(coef) &  coef lower 95\\% &  coef upper 95\\% &      z &      p &  -log2(p) \\\\\n",
       "\\midrule\n",
       "lambda\\_0\\_ & 3067.6037 &  310.1869 &       2459.6485 &       3675.5588 & 9.8863 & 0.0000 &   74.1495 \\\\\n",
       "lambda\\_1\\_ &  153.3781 &   29.9346 &         94.7075 &        212.0488 & 5.0904 & 0.0000 &   21.4161 \\\\\n",
       "lambda\\_2\\_ & 1038.9781 &  168.4720 &        708.7791 &       1369.1771 & 6.1611 & 0.0000 &   30.3667 \\\\\n",
       "\\bottomrule\n",
       "\\end{tabular}\n"
      ],
      "text/plain": [
       "<lifelines.PiecewiseExponentialFitter:\"PiecewiseExponential_estimate\", fitted with 287 total observations, 129 right-censored observations>\n",
       "   number of observations = 287\n",
       "number of events observed = 158\n",
       "           log-likelihood = -1313.4569\n",
       "               hypothesis = lambda_0_ != 1, lambda_1_ != 1, lambda_2_ != 1\n",
       "\n",
       "---\n",
       "               coef   se(coef)   coef lower 95%   coef upper 95%      z      p   -log2(p)\n",
       "lambda_0_ 3067.6037   310.1869        2459.6485        3675.5588 9.8863 <5e-05    74.1495\n",
       "lambda_1_  153.3781    29.9346          94.7075         212.0488 5.0904 <5e-05    21.4161\n",
       "lambda_2_ 1038.9781   168.4720         708.7791        1369.1771 6.1611 <5e-05    30.3667\n",
       "---\n",
       "AIC = 2632.9137"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 261,
       "width": 377
      },
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "naf = NelsonAalenFitter()\n",
    "ax = naf.fit(raw_df['T'],raw_df['event']).plot()\n",
    "\n",
    "from lifelines import PiecewiseExponentialFitter\n",
    "pf = PiecewiseExponentialFitter(breakpoints=[1440, 1500])\n",
    "pf.fit(raw_df['T'], raw_df['event'])\n",
    "pf.plot(ax=ax)\n",
    "pf.print_summary(4)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 261,
       "width": 377
      },
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "kmf = KaplanMeierFitter()\n",
    "\n",
    "ax = plt.subplot()\n",
    "\n",
    "for name, df_ in raw_df[['president','event', 'T']].groupby('president'):\n",
    "    kmf.fit(df_['T'], df_['event'], label=name)\n",
    "    ax = kmf.plot(ax=ax, ci_show=False)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAvMAAAILCAYAAABlxBOjAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuMCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy86wFpkAAAACXBIWXMAABYlAAAWJQFJUiTwAABpTUlEQVR4nO3deXiTVfr/8c8hQCkICYuCQiXiCCqiKIgCosUFdNwYF8YZQYrifNFB0VFHXEYKis7ghqioPxfqDLMPKK6jo1AFcQNkFkFRMKWo4IySKtCytOf3R5KaNkmbptme9P26rlxpn+U8Jwmhd07ucx9jrRUAAAAA52mV6Q4AAAAASAzBPAAAAOBQBPMAAACAQxHMAwAAAA5FMA8AAAA4FME8AAAA4FAE8wAAAIBDEcwDAAAADkUwDwAAADgUwTwAAADgUATzAAAAgEMRzAMAAAAO1TrTHUgnY8xnkjpJ8mW4KwAAAMhtXknfWmsPSuVFWlQwL6lTfn5+l8MOO6xLpjsCAACA3LVu3TpVVlam/DotLZj3HXbYYV1WrVqV6X4AAAAghw0aNEirV6/2pfo65MwDAAAADkUwDwAAADgUwTwAAADgUATzAAAAgEMRzAMAAAAORTAPAAAAOBTBPAAAAOBQLa3OPAAAyHE1NTX65ptv9N1332nXrl2y1ma6S3A4Y4zy8vLUsWNHdenSRa1aZc94OME8AADIGTU1NSovL9fOnTsz3RXkEGutqqqqVFVVpR07dqigoCBrAnqCeQAAkDO++eYb7dy5U61bt1aPHj3UoUOHrAm64Fw1NTXasWOHtmzZop07d+qbb75Rt27dMt0tSeTMAwCAHPLdd99Jknr06KGOHTsSyCMpWrVqpY4dO6pHjx6Svv93lg34Fw4AAHLGrl27JEkdOnTIcE+Qi0L/rkL/zrJBUoJ5Y8wFxpgHjTHLjDHfGmOsMWZBgm31MsY8ZYz5whizyxjjM8bMMcZ0TkZfAQBA7gpNdmVEHqlgjJGkrJpUnayc+VslHSVpu6TNkg5NpBFjzMGSVkjaT9JiSR9JGiJpqqTTjTHDrbVfJ6XHAAAAQBOEgvlskqyPrddK6iupk6QrmtHOPAUC+auttWOstdOstSdLul9SP0mzmt1TAAAAIEckJZi31i611n5im/GdQ3BUfpQkn6SH6+2eLmmHpPHGGJLgAAAAAGVXacqRwftXrbU14Tustd8ZY95SINg/XtLrDTVkjFkVY1dC6T9JkYVfy4Tb3bat3jn9dK0eObLxgxX4mmnffffVAQccEPc1XC6XvF6vCgoKEu0mAABwoNLSUo0cOVLTp09XcXFxpruTU7Jpdki/4P36GPs/Cd73TUNfWpy2u3fr+Fdeift4a63++9//Nuka1dXV8vl8TewZAABoKmNMxC0vL09er1cTJkzQunXrMt3FpFqwYEHt43ziiSci9vv9ft199926+OKLdfjhh6t169Yyxui1117LQG+TK5tG5t3B+4oY+0PbPY01ZK0dFG17cMT+mCb3rIVo28QyS4lkVVVXVzf5HAAAkJjp06fX/lxRUaH33ntPv/3tb7Vw4UItX75cAwcOzFznkqS8vFxTpkzRPvvso+3bt0c9xufz6Ze//KUkqVevXurWrZu2bt2azm6mTDYF87mtkcC3tLS09ufCwsLU9iXsmj6fT0UTJ9Zuu6ahN3XXrtKAASopKandFG9fwx8fAABIj2gpLVdddZUeeughzZkzp87fdCey1mrixInq2rWrzjvvPN1zzz1Rj+vdu7dee+01HX300erSpYuKior09NNPp7m3qZFNaTahkXd3jP2h7f7Ud6UFW7Mm9u1rqoICAOB0o0aNkqSIdNni4mIZY6IOwPl8PhljVFRUVGf71q1bdf3116tfv37q0KGDPB6P+vXrp6KiIm3cuDHq9desWaMzzzxTHo9H7du310knnaQVK1Yk9Fjmzp2rJUuWaP78+Q0uFNa5c2edcsop6tKlS0LXyWbZNDL/cfA+Vk78IcH7WDn1SIZYI/Nr1qSzFwAAIEVCeeKDBw9uVjs7d+7U8OHDtWHDBp122mk6++yzZa1VWVmZFi9erAsuuEB9+vSpc87KlSs1e/ZsDR06VJMmTdKmTZu0cOFCnXLKKVqzZo369esX42qR1q1bp2nTpmnq1Kk68cQTtWTJkmY9HqfKpmB+afB+lDGmVXhFG2NMR0nDJe2U9E4mOgcAAOA04Wk23377rd5//3299dZbOuuss3T99dc3q+3XX39dGzZs0DXXXKP777+/zr7du3drV5S5eC+++KLmz59fZ4T/scce0+TJk/XAAw9o3rx5cV177969Gj9+vA488EDdeeedzXocTpf2YN4Y00bSwZL2WGs3hLZbazcYY15VoPzkzyU9GHbaDEkdJD1mrd2Rzv4CAIDc4aQ5XMmYQzdjxoyIbYcffrh+8pOfqGPHjs1uX5Ly8/MjtrVt21Zt27aN2D58+PCIVJ1LL71UU6ZM0XvvvRf3NWfOnKkPPvhAy5cvj3r9liQpwbwxZoykMcFfewTvhxpjSoI//89aG/r411PSOkllkrz1mrpS0gpJc40xpwSPO06BGvTrJd2SjP4CAAC0BOGV53bs2KEPP/xQ06ZN08UXX6wPP/xQs2bNSrjtk046ST179tSvf/1rrV69Wj/84Q81fPhwDRw4UC6XK+o50VJ72rRpo+7du2vbtm1xXffdd9/VnXfeqeuuu05Dhw5NuP+5Ilkj8wMlTai3rU/wJgUC90a/ywmOzg+WNFPS6ZJ+KOlLSQ9ImmGtje9VRsJKY9WBr6qSfD6ppET+LVtqN5fMmRNXu/6qqtqffQnk3xtj5MnLk7tdu+835udL3bs3ua1YWNQKAJDLOnTooCFDhmjRokXq1auXZs+ercmTJyf8d69Tp0565513NH36dD333HN6JbheTbdu3XTllVfq1ltvVZs2beqc4/F4orbVunXruMpX7927V5dccon69u2r22+/PaF+55qkBPPW2mJJxXEe65MUczlUa225pImx9iN5on1q9vn90Q+2Vtq8WZJUFfZm8+/eHde1qsJGBrY0sZ69FPiHus0YeVqFFWDaZx+psrLB84wx8ng8crtjFUn6XmhRK4J5AMhd6Sr/nM1CFWdWr16t1atX1/7daxX8G7t3796Ic/wx4oNevXrpySeflLVWa9eu1ZIlS/Twww9r5syZqqmpSXrAvX37dq1fH6iF0i58gC/M5Zdfrssvv1xTp07VnDgHHZ0smybAIs28Xm/8K7KG5aO13rlTe2uC85Nbx/lPKMp/DE2xV4FP7Qq9cWMsClGftVZ+vz+uYF5iUSsAQMsQSmmpqamtN6LOnTtLCizCVN/KlSsbbM8Yo/79+6t///4aM2aMDjzwQD377LNJD+bz8vJ02WWXRd23evVqffDBBzrhhBPUr1+/FpOCQzDfghUUFESMUHhjfP0VS7zHV1RVyb9rV0KrxoZSdDzt2n1/vaoqyeORvN642mhsJMZJE6IAAGiOZ599Vp999pnatGmjYcOG1W4fMmSIJGn+/PkaP358YBBNgeB+5syZEe18+OGH6tatm7rXS3kNrazavn37pPc9Pz9fTzzxRNR9xcXF+uCDDzRhwgRNmjQp6dfOVgTzLVz9lJLCOILj8Lz6eI5vjlKfrzb1x+vxfH89vz8QyDcQpBOgAwBauvDSlDt27NDatWv18ssvS5LuvPPOOoH4cccdpxNPPFFvvvmmhgwZopNPPllbt27V888/r9GjR0eM2P/jH//QDTfcoKFDh6pv377ab7/9tHnzZi1evFitWrXSDTfckJbHGK/rr79e//vf/yRJy5cvlyTdfffdWrBggSRpzJgxGjNmTKa6lzCC+SyU7iC0MPzacaTdhOfVx3N8c/j8/tqReZ/f//31QhNyS0uZuAoAQAzhpSldLpf23XdfnX322ZoyZYpOO+20iOMXL16sG264QYsXL9aDDz6oQw45RLNnz9aoUaP0l7/8pc6xo0eP1qZNm/Tmm29q8eLF+vbbb7X//vvrtNNO0y9+8Ys6o/7Z4G9/+5vKysrqbHv11Vdrf/Z6vQTzSJzL5XJkvnZZRUVklZk0Y+IqAAB1JZLWKgUmxz7++ON6/PHHG23zsMMO03333RdXu4WFhQ32Ke45fA0oLi6u801EKq6RjQjms0RoMmqmA/rCibELCe1t106+c89V2dChtW9Ia638u3ZlLpgPvjGrpUAO/YABmekHAABABhDMZ4mCgoLMjSzn5zda4lGSWldV6QfPPae8n/xEPr9f1WGfsFOVOx8zZ76mRvL7VRpWv15ff52ca5K6AwAAHKJV44cg5xUVSXl58R1bWakCt1sjevdOaZca1aePNHCg1KNH4NZM9Wvuh1J3AAAAshkj85DGjpX69g0Ex7GMHJm27jSk/oTb2sm4e/eq1OeTa9myhEbUo6U5ZTrlCQAAoDEE88h6xhhZa6OuThuqdKOamsD+DRu0ceNG9e7du87IejwVglwulw4++GBt2LAhKf0GAABINYJ5ZDWXMfLk5TVpwalox8VKmTHGyOPxyO12k1oDAAAch2AeWS204musajnhaTb1hUb0G2Ktld/vl9vtlkRqDQAAcBaCeWS1ArdbBcFAO5o6i0h5PIFVYRWoZ1teXp4V5T4BAABShWAeOauxcp/pXmkXAAAg2QjmkVtCI/XxBOrR8uPz86Xu3ZPYIQAAgNQhmEduyMuTtmz5/vcolW8ihC84FVKvZn0io/csOAUAANKFYB65oXPnur/HsyJt/ZH54IcBl8vVrDz7UFUcgnkAAJBqBPPISfUXl4rG5/fLSPK0a1enWk60BaSaikm3AAAgHQjmkRTxBM/1uYyR1+NpsFpNU5RVVMRdi176fsGpbVVV8rRrFyhvGVx4KlFNXagqFlJ1AAC5pLS0VCNHjtT06dNVXFyc6e7klFaZ7gCcy2VMs86vjrGqa6J98OTlyTSzT1IgII91KysrU0VFRbOv0RgWsAIANIcxJuKWl5cnr9erCRMmaN26dZnuYlItWLCg9nE+8cQTEfv//e9/a9KkSTr66KO17777Ki8vTwUFBTr11FO1aNGiJg0GZhtG5pEwr8cjn9+v6ma8AZpzbv0+uOulyzRmTdiEWWOM4ulJ/UWmUolUHQBAc02fPr3254qKCr333nv67W9/q4ULF2r58uUaOHBg5jqXJOXl5ZoyZYr22Wcfbd++Peoxq1at0rPPPqvjjz9ew4YNk9vt1pYtW/T888/r/PPP1/jx4/Xb3/42zT1PDoJ5NN3IkZKkguAtY/LzVVBUpIKxYxM6PfxbAU9envzffRc1oDfGyOPx1AngCwsLo7YZnloT65jGUP8eAJAs0VJarrrqKj300EOaM2eOSkpK0t6nZLLWauLEieratavOO+883XPPPVGP+8lPfqKioqKI7d9++62OP/54/e53v9OUKVM0ZMiQFPc4+UizQXzy8zPdg0iVlVIz/hMKT8lxt2un3q1by+vxyOv11rn17t1bXbp0SUKHAQDIvFGjRkmS/vvf/9bZXlxcLGNM1EEln88nY0xEQLx161Zdf/316tevnzp06CCPx6N+/fqpqKhIGzdujHr9NWvW6Mwzz5TH41H79u110kknacWKFQk9lrlz52rJkiWaP3++OnToEPO4vLy8qNs7deqk0aNHS5I++eSThPqQaQTziE9RUfYG9AlqSo49KS8AgFzx2muvSZIGDx7crHZ27typ4cOH695771Xv3r11xRVX6LLLLtOAAQO0ePFirV27NuKclStXatiwYaqqqtKkSZN01llnafny5TrllFP08ccfN+n669at07Rp0zR16lSdeOKJCT+GJUuWSJIGDBiQUBuZRpoN4jN2bOCWJOHVbwrjqQlfXzDVpznCc+wLvd5AJZt6fYlZpSdWKkx4NZvGvjWIsdpsYxVxqHQDAIhXeJrNt99+q/fff19vvfWWzjrrLF1//fXNavv111/Xhg0bdM011+j++++vs2/37t3atWtXxDkvvvii5s+fX2eE/7HHHtPkyZP1wAMPaN68eXFde+/evRo/frwOPPBA3XnnnXH3+dNPP9WCBQtUXV2trVu36sUXX9QXX3yhm266SUceeWTc7WQTgnkgpFMnac2autuirRIrxVxh1lVVpbjH8OutNhsvFqUCgMQ5aV5SonOvws2YMSNi2+GHH66f/OQn6tixY7Pbl6T8KN/ct23bVm3bto3YPnz48IhUnUsvvVRTpkzRe++9F/c1Z86cqQ8++EDLly+Pev1YPv300zrPSdu2bXX33Xfruuuui7uNbEMwDyg4At+qleTx1Nkeq3RmqccTtU6+t6Iivgo/YZV0EkHaDwAgHuElF3fs2KEPP/xQ06ZN08UXX6wPP/xQs2bNSrjtk046ST179tSvf/1rrV69Wj/84Q81fPhwDRw4UC6XK+o50VJ72rRpo+7du2vbtm1xXffdd9/VnXfeqeuuu05Dhw5tUp9PP/10WWu1Z88ebdq0Sb///e91880364033tDChQujfgDJdgTzaLFcxjS7rKbP768TzBe43fEtghVK6Yky6tJQRRwnjSgBALJLhw4dNGTIEC1atEi9evXS7NmzNXny5IS/6e3UqZPeeecdTZ8+Xc8995xeeeUVSVK3bt105ZVX6tZbb1WbNm3qnOOpN2gW0rp167gGqvbu3atLLrlEffv21e23355Qv6XAB4iDDz5Yt912m9q2baubbrpJc+fObXbqUSYQzKPFyoY6+QCA9EpG6orThSrOrF69WqtXr64N5lu1CtRF2bt3b8Q5/hjfVPfq1UtPPvmkrLVau3atlixZoocfflgzZ85UTU1NswLuaLZv367169dLktrFWFvm8ssv1+WXX66pU6dqzpw5jbZ5xhln6KabblJpaSnBPOAk8Yyix5oA6/P7ZSR5mrBIFQAA2SKU0lJTU1O7rXPnzpICizDVt3LlygbbM8aof//+6t+/v8aMGaMDDzxQzz77bNKD+by8PF122WVR961evVoffPCBTjjhBPXr1y/uFJzPP/9cUuDbASdyZq+BDKiflmMl+WNNkAUAIEs9++yz+uyzz9SmTRsNGzasdntowaT58+dr/PjxtcFteXm5Zs6cGdHOhx9+qG7duql7vcpsW7dulSS1b98+6X3Pz8/XE088EXVfcXGxPvjgA02YMEGTJk2qs2/lypVR8/X/+9//atq0aZKkM888M+n9TQeCeSBO0dJySLIBAGSz8NKUO3bs0Nq1a/Xyyy9Lku688846gfhxxx2nE088UW+++aaGDBmik08+WVu3btXzzz+v0aNHR4zY/+Mf/9ANN9ygoUOHqm/fvtpvv/20efNmLV68WK1atdINN9yQlscYj0mTJunrr7/WkCFDdOCBB8rlcsnn8+mll15SZWWlxowZo0svvTTT3UwIwTwyLmYt96BoVWMyYUPYLPvwEflY/c+WfgMAWq7wMowul0v77ruvzj77bE2ZMkWnnXZaxPGLFy/WDTfcoMWLF+vBBx/UIYccotmzZ2vUqFH6y1/+UufY0aNHa9OmTXrzzTe1ePFiffvtt9p///112mmn6Re/+EWdUf9Mu/766/Xss89q9erVeuWVV7R7925169ZNJ598ssaPH6+xY8fGvZBktiGYR0Y0pZJMtKox6dKcijep7nesyjYsKgUAsAn+7fJ4PHr88cf1+OOPN9rmYYcdpvvuuy+udgsLCxvsk6+Rgb14FBcX1/kmIty4ceM0bty4Zl8jGxHMIyOaWkkmU1VjmlvxptHzogXkof/QoiyC4XK5Gi3dxaJSAAC0HATzyIh467E3loKTarH6GWsxqWj7wx9DbepNtNVmQ0IpPD161Bl9d7lcys/PV2VlZVwBPQAAyH0E83C+kSMD9/n5UlGRNHZsyi9pjEnoK8za1Js+fWIe4yorU/WXX0aeW12tyspKjRgxIua5LCoFAEDL0irTHQASEiUFRZWVUklJWi7vyctLeKJMY6k3Xo9H0RfBZsQdAADUxcg8nKmoKBC4V1bW3V7/9xRxt2snd3DBqEKvN2J/eGpNaH+8KUMFbrcK2rWTvF4puFIhI+4AACAagnk409ixddNpQqk2GVA/SHcZo4qqqtpgHwAAIFVIswES4GogxabaWlaGBQAAaUEwDyTA6/E0GNCzMiwAAEgH0myABMQqWZnpUprhmppnz2JTAAA4D8E8kGKhAD9W7XkprP58M1eLjWdRqVhYbAoAAOchmAdSoLGyldEWndro96t3KJivqgqsBBscXQ9f5nrZsmUxR9C9Xq98Pl+zAnoAAOAcBPNACnjy8uQypk5NeaOGc+mjLkIVCuLDgv/qL76QT4oazBcUFCQ0sk7pSwAAnIlgHo7RUD56YSPHJSuNJV7udu00onfvOtvKKyrk8/sjFo2KNkqvvDxpy5bvf9+79/ufKysZQQcAAJII5pHl6o9uJ6raWvn8/rQF89HEM2m2dgGqegtR1ebdb96cms4BAABHojQlslpjJSCbIhkfCgAAcJqPPvpIV111lY444gi53W61bdtWBxxwgM4880w9+eST2rVrV8quXVJSImOMSkpKUnaNlo6ReWS1WKPZDSmMMaqdbqmqWAMAQLxmzpypGTNmqKamRkOHDtWECRO0zz77aOvWrSotLdWkSZP0yCOPaOXKlZnuKhJEMA8kUUNpQdmQ6gMAaDnuvPNOTZ8+XQUFBfrrX/+q4447LuKYF154Qffee28GeodkIc0GSKLG0oKckOpTWloa123ZsmUqLy/PdHcBAFH4fD4VFxerTZs2eumll6IG8pJ01lln6e9//3vt7yUlJTr//PPVp08f5efnq1OnTho+fLgWLFgQ9fzCwkIZY7R7927NnDlT/fr1U15enoqKilRYWKiJEydKkiZOnChjTO0tvOTy3r17NW/ePB1//PHq1KmT2rdvr6OPPloPPfSQampqIh6XMUZFRUVav369fvzjH2u//fZTq1atWmxlNkbmkXtGjqzza2FTzs3Pl4qKpLFjE7p0OlaG9fn9MmVlKi8vT9oCT4ksNsUiUwCQvebPn689e/booosu0hFHHNHgsXl5ebU/X3HFFerfv79OPPFE7b///vr666/10ksvafz48fr44491++23R23j/PPP1/vvv68zzjhDY8aM0X777afCwkJ5PB4tXrxY5557rgYOHFh7vMfjkSTt2bNHZ599tl555RX169dPP/3pT9WuXTstXbpUV111ld5991397ne/i7jehg0bdNxxx6lv3766+OKLVVlZqU6dOjX9icoBBPPIDfn5UmVl89uprJRKShIO5lOlfvqOtVZLlixR79695XK5Yi4iFa9EF5uiRCYAZKfly5dLkk455ZQmnfef//xHBx98cJ1tu3fv1hlnnKFf//rXmjx5snr27BlxXllZmf7zn/+oW7duEfsWL16sMWPGqKioKGLfrFmz9Morr2jKlCmaM2eOXC6XpMDfl5/97Gd66qmndMEFF+jcc8+NeHw33XST7rzzziY9vlxEMI/cUFQUCMKTFdBnyLKysqiTZL0eT2BEXt8vPBVaZCoZI+RNXWyqpX6VCcD5nPT/V2FhYcLnfvnll5KkXr16Nem8+oG8JLVt21Y///nPtWTJEr3++uu65JJLIo65/fbbowbyDampqdGDDz6oHj166P77768N5KXAN8b33nuv5s+fr9///vcRwXz37t01ffr0Jl0vVxHMIzeMHRtzND1qHfdo6qXnpEv4qHusSbKh9B3vf/+rJcZErCTLCDkAIBk2bdqk3/zmN3r99de1adMmVdYb4Pr888+jnjdkyJAmX2v9+vX65ptvdMghh+iOO+6Iekx+fr7WrVsXsf2oo46qkx7UkhHMAxkWGnUPD+hjKWjdWr3d7ohFpQAACLf//vtr3bp1MYPvaDZu3KghQ4Zo27ZtGjFihEaNGiW32y2XyyWfz6enn346Zk36Hj16NLmPX3/9tSTpk08+0YwZM2Iet3379qRcL1cRzAMZFhp1b9IkWZ8vME+ge/dUdQsAclJzUlec5IQTTqhNi7nsssviOue+++7T119/rfnz50fkt//xj3/U008/HfNck8ACj+7gt9A/+tGPtGjRoiadm8j1chWlKQEn6dRJ2rIlcMtgbj8AILtNnDhRbdq00cKFC7V27doGjw2Ntn/66aeSApVp6nvjjTcS6kf4hNb6Dj30UHk8Hr3zzjvas2dPQu2DYB5Iu1KfL+rN5/fX3mIds8zlUkWwnBcAALF4vV4VFxdr9+7dOvPMM2Ou8Pr3v/9dZ5xxRu05UuQk4VdeeUVPPPFEQv3o2rWrpEAufn2tW7fWVVddpS+//FJXX311RH6+FJjI29iHkZaONBu0KA2lshTWO85lTNTKMoloaGXYpqi2Vv5du8QasgCAxtx8883au3evZsyYoWOPPVbDhg3T4MGDtc8++2jr1q1688039cknn2jw4MGSpCuvvFLz58/XhRdeqAsuuEAHHHCA/vOf/+jvf/+7xo4dqz//+c9N7sPQoUPVvn17zZkzR19//XVtrvtVV10lt9utX/3qV/rnP/+pRx99VM8//7xOPvlk9ezZU1999ZU++eQTvfXWW5o1a5YOP/zwpD43uYRgHjkv0UA6VmWZRNSf5NocNstWkV22bFmz69wDAFLjtttu04UXXqh58+Zp6dKlmj9/vqqqqtS1a1cNHDhQN954o8aNGydJOvLII7V06VLdeuutevHFF7V3714dddRRWrRokTweT0LBfOfOnbVw4ULNmDFDJSUl2rFjhyRp3LhxcrvdatOmjZ599lktWLBAJSUleuGFF7R9+3btu+++Ouigg3T77bfr4osvTupzkmsI5pHzmhNIJyP4lmKvDBuusRKayVxFtrnCV4xlJVgAyG6HHXaYHnzwwbiOHTZsmJYsWRJ1X7TBpHjq9p9++uk6/fTTY+43xmj8+PEaP358o215vd6sG9TKNIJ55Lx4AulckK6FUFwul/Lz81VZWVknoAcAAOnHBFjAwcJXy0uX6upqVVZWasSIEWm/NgAAqIuReSALRUup8fn9kiR/VZV8e/dKfr86f/ON/H5/na8cjTHyeDy19XtTgZF4AACyA8E8kCUSmajrdrujBu0ulyslI+fpSuUBAADxIZgHoiicODH1F8nPl4qKpLFjJSW34g0j5wAAtAwE80BIfn56V1WtrJRKSmqD+cYm6oZSb3x+v7xVVZLHE7EsOSPnAAC0LEyABUKKigIBfTql88MDAADIOYzMAyFjx9aOkjdW873ZRo5s1ukVNTVyK3Ik3ufz1U6ABQAAuY9gHnCI0ARZI8kfDOajsdbKH6x8k07NSfFxuVysIgsAQAKSlmZjjOlljHnKGPOFMWaXMcZnjJljjOncxHZOMMYsDp5fZYzZZIx5yRgTe+kwoAXwBkfbPe3aNXpsOlbHKy0tVVlZmXw+X+2trKxMFRUVTW4rtIosAABomqSMzBtjDpa0QtJ+khZL+kjSEElTJZ1ujBlurf06jnaukDRP0g5Jz0jaLKmXpPMknWGMudVaOysZfQacpsDt1oZt2+Ru107u1q0D6T/1JsCWlJSktA8ul6tOpRyPx1Onzn3oW4FEatxTgQcAgKZLVprNPAUC+auttQ+GNhpj7pN0raRZkiY31IAxpo2kuyRVSRpkrf04bN+dkj6QdIsx5h5r7a4k9RtAE3i9Xvl8vtrAO1ad+/pVdhpCBR4AABLX7GA+OCo/SpJP0sP1dk+X9DNJ440x11lrdzTQVBdJbkn/Cg/kJclau84Ys17SAEn7SCKYBzKgoKAgZl47QTkAAOmXjJz5UFmOV621NeE7rLXfSXpLUntJxzfSzleS/iuprzHmkPAdxpi+kg6RtCaedB0AAAA0TagiWlFRUZ3tRUVFMsYwtylLJSPNpl/wfn2M/Z8oMHLfV9LrsRqx1lpjzM8lLZC0yhjzjKQvJPWU9CNJH0q6KJ4OGWNWxdh1aDznA+FKU/CfV2GC7ftCVWr27g2cV280PLyKTXl5OdVhAAD66KOP9PDDD2vp0qUqLy9XZWWlunXrpqOPPlrnnXeexo0bp7y8vJT3o6SkRBMnTtT8+fMjPjAgcckYmQ8lzMYqYRHa7mmsIWvtXyWdLMkv6RJJ0ySNV2BC7HxJG5vRTyBuLmMy3YVmYwQFADBz5kz1799fDz30kDp16qQJEybo+uuv1xlnnKGPPvpIkyZN0vDhwxts46677tK6devUs2fPNPUaTZFVdeaNMeMkPS5pkaTbJZVJ6i3pV5IeknSSpLGNtWOtHRSj/VWSjklWf5G7vB6PfH6/qtNQ4jFVqA4DAC3bnXfeqenTp6ugoEB//etfddxxx0Uc88ILL+jee+9tsJ39999f+++/f6q6iWZKRjAfGnmPVYsutN3fUCPBvPinJP1L0viw/PuPjDHjFUjnudAYU2itLW1Wj4FGFLjdKkigvGIimrLCbG1KTlWVDu7cWQX1qsb4fL6MLBgFAMguPp9PxcXFatOmjV566SUdccQRUY8766yzdNpppzXYVlFRkZ5++ml99tln8gb/Zvl8Ph100EGaMGGCiouLNW3aNL322mvavn27jjjiCBUXF+uss86qbaOwsFBvvPGGJGnixImaOHFi7b7wdisqKvTrX/9aixYtUllZmfLz8zVkyBDdcMMNOvXUU+v0q7S0VCNHjtT06dM1ZswY3XLLLXrrrbe0e/duHXvssbrrrrs0bNiwpj51jpOMYD5UeaZvjP2hyayxcupDRklqI+mNKBNpa4wxb0oaFLyVJtZVwNlCq8BKgfz5gvoVZPx+qaoq8HOUnPqUCkvrKW1CvXtf2IePp9eskaddO7njWBgrp+XnS927p/2yrMQL5I758+drz549uuiii2IG8iHNyZcvKyvTkCFD1KdPH40fP17ffPON/vznP+vcc8/Va6+9ppEjA3VSioqK5PF4tHjxYp177rkaOHBgbRue4KKIfr9fw4cP19q1a3Xsscfqmmuu0f/+9z/95S9/0ahRo/TII4/o//7v/yL6sHLlSs2ePVtDhw7VpEmTtGnTJi1cuFCnnHKK1qxZo379+kWck0uSEcwvDd6PMsa0Cg/EjTEdJQ2XtFPSO420E/qXtG+M/aHtuxPtKOB0Xo9HG7Ztk/LyVP3ll9K2bXUP2L5dqgm+BbdsCQT3aeKqqlIiiT1m716FkpmsJP/27XK3zqoMwPTr0SMjlw2txEswDzjf8uXLJUmnnHJKSq9TWlqq4uJiTZ8+vXbbT3/6U51++um6++676wTzkrR48WKNGTMm6gTYG2+8UWvXrtXPfvYzPfroozLB+Ws33nijBg8erKuvvlqjR4+uHcUPefHFFyMm1T722GOaPHmyHnjgAc2bNy+pjznbNPsvprV2gzHmVQVG1n8u6cGw3TMkdZD0WHiNeWPMocFzPwo7dlnw/oLgwlD/Cjt+oKQLFPhbv6S5fQayysiR0bfn50tFRdLY76eJhFaBVefOgQ31U3TWrPl+ZL5Hj8j9KeStqEhonoGnqkr+qqo6Ab2CozQt0pYtGb08cy2Q65y0JkZTFuCr78svv5Qk9erVK0m9ia5379669dZb62wbPXq0DjzwQL333ntxt7N7924tWLBA++yzj+66667aQF6SDjnkEF199dW644479Nvf/la33XZbnXOHDx8e8eHg0ksv1ZQpU5rUB6dK1vDXlZJWSJprjDlF0jpJxylQg369pFvqHb8ueF/7Sllr3zPGzJc0UdL7wdKUZZK8ksZIaitpjrX2wyT1Gcic/HypsrLhYyorpZKSOsF8NmvuPIPwEp1NmUeQc/z+wIewZvwRT4STAhwA2WPgwIFyuVwR2wsKCvT222/H3c7HH3+snTt3avjw4erSpUvE/pNPPll33HGHPvjgg4h9gwcPjtjWpk0bde/eXdvqf4Odg5JRmlLW2g2SBksqUSCIv07SwZIekHR8ExZ6ukyBYP5tSaOD7Zwmabmkn1hrr01Gf4GMKyoKBPSNaSzgBwAgilD1mc8//zyl1/HE+Ca1devWqqmpibovmoqKQD2VWFVzQtujFXloqA8t4dvGpCWmWmvLFQjE4zk2ahFva61V4ANBSbL6BWSlsWMbHnGPlXoDAGiW5qSuOMkJJ5ygJUuW6PXXX9dll12W6e40yh38ZndLjFTDUNqQO02V5pwkKSPzAAAAyB4TJ05UmzZttHDhQq1du7bBY3ft2pWWPoXScaKNlvfr10/t27fXP//5z6ij70uXBuqtHHMMywXV18JLRgC5y+f318lDz3bhJSob6rfLGHk9nrStAwAATuT1elVcXKxbbrlFZ555pv76179GzS3/+9//rtmzZ2vJktTXF+nataskadOmTRH72rZtq4svvliPP/64fvWrX+nBB7+vp7JhwwbNnTtXbdq00fjx41PeT6chmMf31qxpeH+nTlKfPmnpChITPvvfX1VVJ0DOJkZKuJ58tbWBGvsE8wDQoJtvvll79+7VjBkzdOyxx2rYsGEaPHiw9tlnH23dulVvvvmmPvnkk6hBfioMHTpU7du315w5c/T111+rR7AM71VXXSW3261f//rXWrZsmR566CG9//77GjlyZG2d+e+++04PPfSQDjrooLT01UkI5iF17SqFLd4QU2PBPtKu/gh21Z49qtq7t/Z3f6hMZRO0a91a7VJc590q0LdEF4dqavlLAGipbrvtNl144YWaN2+eli5dqvnz56uqqkpdu3bVwIEDdeONN2rcuHFp6Uvnzp21cOFCzZgxQyUlJdqxI1C1fNy4cXK73erSpYvefvtt3XXXXVq0aJHuu+++OivAjho1Ki39dBpjW9AfRWPMqmOOOeaYVatWZborzlNaGgjm4wn60XzhE2CXLq2za1lZWcxgtqKqSp+FjcZ7EgiWjTHqnaZR7/ASlOEfTA7u3DnqyHuLKF8Zep9lsDRlS5kgiNy0bl2g+vVhhx2W4Z4gV8X7b2zQoEFavXr1amvtoFT2h5F5wGG8Hk/MxZnc7drVCeC9CS6+lMpAOVY+vMuY2sdEGg0AAPEhmAccprHFmRIdvc70ZFmvxxNY3Vak0QAAEC9KUwLICozEAwDQdIzMAzmsKaPt8ZaGDEeZSAAAMotgHsgx4bnnqUaZyNwUPhlWCiz04vV6VVBQkJkOAQBiIs0GyDFej0eusHrzqUZ+e24IrcwYTXV1tXwOWoAMAFoSRuaBHNPYBNlYmjpxNtMTZpFcXq9XPp8v6jLrUvTl1wEAmUcwD2S78JrzkpSfLxUVSWPHZqQ7yE0FBQVR02jqp9wAALILwTyQjfLzpcrK6PsqK6WSkpQG8w2NuocmvWbasrIyJt8CAFo8cuaBbFRUFAjoY4kV6DdDvHn2oUmvmRDex0z2AwCAbMHIPJpmzZpM9+B7nTpJffpkuhepMXZs9JH3+ik3SdTQyrL1ZWrSa/0+MvkWANDSEcwjPl27SgMHZroXdWXTB4scEM/E2UxPeg31MdP9AAAgWxDMIz4DBmS6B3UxKQ8AAICceQAAAHyvpKRExhiVlJRkuiuIA8E8AABADlu5cqUmTpyoPn36KD8/X506ddKAAQN0ww036PPPP89099BMBPMAAAA5yFqrG2+8Uccee6wWLFigQw89VFdffbUuu+wytW/fXvfcc4/69u2rv/3tb5nuKpqBnHkAGVPq89XWradePAAk1+23367Zs2fL6/XqhRdeUP/+/evsX7hwocaNG6eLLrpI//jHPzQyhRXTkDqMzANIq/r17KkXDwDJ5/P5dPvtt6tNmzZ67rnnIgJ5STr//PN1//33q7q6WldccYVqamoijnnxxRc1bNgwdejQQZ07d9YFF1ygTz75JOK49evXa9q0aRo8eLD23Xdf5eXlqXfv3vrZz36mzZs3RxxfWloqY4yKi4u1cuVKnX766XK73ercubPOP/98lZeXS5I2btyoiy66SPvuu6/y8/M1cuRI/fOf/2z29XMJwTyAtPJ6PFEDegBA8syfP1979+7Vj370Iw1ooCLdpEmTtP/+++vjjz/WG2+8UWffokWLNGbMGPXq1UtTp07V0KFDtXDhQh1//PH6+OOPI4599NFHVVBQoJ/85Ce66qqrdPjhh+uJJ57QscceGzM3//3339eIESMkSZdffrmGDBmiRYsW6dRTT9VHH32kIUOGaPPmzbrkkkt05pln6o033tBpp52m7du3J+X6uYA0GwAJCR9NT7Tue/02Qr8bY1ReURFX6k1zas6T4gMgVy1fvlySdOqppzZ4XOvWrTVy5Ej94Q9/0FtvvVUn1eb555/X888/r7POOqt22wMPPKBrrrlGV155pV5//fXa7ePHj9e1116rvLy8Ou2/+uqrOuOMM3THHXfokUceibj+Sy+9pAULFujiiy+u3XbZZZfpqaee0rBhw3Tdddfplltuqd13++2367bbbtOTTz6pqVOnNvv6uYBgHkDcXMakZRTdWqslPp96xwiyyyoqZBvoh5HkaddO7nbtGrxOKMWHYD4+5eXlKigoyHQ3gGYpddA6JYWFhQmf++WXX0pSXO/Z0DFffPFFne0nn3xynUBekqZMmaIHH3xQS5YsUVlZmXr37i1J6tmzZ9S2R40apf79++uVV16Juv+EE06oE8hL0oQJE/TUU0/J7XZr2rRpdfZdcskluu2227Sm3sKRiV4/F5BmAyBu0VJkkim85YaCdU9enkwD/bCS/FVVcV2TFJ+GuVyu2p99rLwLtCgnnXRSxDaXy6UTTjhBkvTBBx/UbrfWasGCBTr11FO17777qnXr1jLGyBijf//73zHTXAYPHhyx7YADDpAkDRw4sM7/QdL3QXv9PPhEr58LGJkHnOqHP5SKiqSxY9N2yQK3u3YUO570lsbSWMLbKPR6VV5RoSWffabGwmt3HKPuoTZjaU56Tkvi9Xq1YcMGSVJ1dXWGewMgXj169NC6detqJ5I2JHRMKIgO6d69e8y2JamioqJ22y9+8QvNmTNH+++/v0aPHq2ePXsqPz9fUmARqrKysqhtuaP8fWjdunWj+/bs2VNne6LXzwUE84CT5OdLlZWBnysrpZKStAbz4eJJuWlqGkuB263eHk/t7w0F47EQpCdXQUFBbTAP5ILmpK44yQknnKClS5fqtdde0+WXXx7zuOrq6trUo+HDh9fZt3Xr1qjnbNmyRdL3wfZXX32luXPn6ogjjtCKFSvUsWPHOsf/8Y9/TPRhxCXT18800mwAJykqCgT0IaHAPgPiTbkhjQUA0q+oqEgul0vPPPOMPvzww5jHPfXUU/riiy/Ur1+/iLSa+tVtpEDwH5pce/TRR0sKlI+sqanRqFGjIgLpzZs3a+PGjc19OA3K9PUzjWAecJKxY6WXXsp0LyQFRtFH9O6tQq836g0AkDl9+vTRzTffrD179uicc87R2rVrI4559tlnNXXqVLlcLj3yyCNq1apuWLhkyRK98MILdbY99NBD2rBhg0aOHFk7+dUb/D9/+fLlddLxtm/frssvv1x79+5N8qOrK9PXzzTSbAAAAHJQcXGxduzYofvuu09HHXWURo8erf79+2vPnj1asWKF3n33XeXn5+uPf/xj1NVfzz77bP3oRz/Sj370I/3gBz/QmjVr9PLLL6tLly6aN29e7XE9evTQRRddpD/96U8aOHCgRo0apYqKCv3jH/9Qu3btNHDgwIjqM8mU6etnGiPzcLY1a76/5fjXaAAANEWrVq1077336t1339VPf/pTffjhh5o7d67+3//7f9q+fbuuu+46rV+/XhdeeGHU88877zw988wzKi8v1wMPPKAVK1bovPPO09tvv61DDz20zrFPPvmkbr75ZlVWVurhhx/WK6+8orPOOksrVqyIOpE12TJ9/UxiZB7O1LWrNHBg3W05/KkbAIBEDRkyREOGDIn7+KKiIhUVFdX+Xr/WfDTt27fXrFmzNGvWrIh90Wr7FxYWxixB7PV6GyxPHG1fU6+fSwjm4Uz1l6bO8TdqS5VIZZrGVqYNlcsEACAXkGYDIKukclEq6ftymQAA5AJG5gFkFa/HI5/fn5SSlvEG7SVpTNEykjz1Fr1qbHEtAABiIZgHkFXCV5lNxLKyspgfBMKDe2NMgzmZqWIl+auq6gTzTV1cCwCAENJsAOSUeBez8uTlyaQ4pSeWaB8hWFwLAJAIRuYBZIVEJrs2JFrqSvg1MrGwVbTrJ/txAwBaFkbmAWRMKie7MtEVANASEMwDyJh4U2ISRepK8i1btkzl5eWZ7gYAIIg0GwAZ09zJrrGQupJcLpdL1dXVkqTq6mr5fD4VFBRkuFcAAImReQBAI7xer1wuV+3vocAeAJB5BPMAgAYVFBRoxIgRme4GACAK0mwApNyysrKsWxSpOak4LPIEAMgWjMwDSInwia3ZUlkmWZNts+XxAADAyDyAlPB6PPL5/bUVZbKhskz9PjVHc9sIfTMQ+lBgJJVXVIhppQCApiCYB5ASoUo12VRZJhnVc5qbnhPrQ4BVILAnmAeQDPVXuG7VqpXcbreOPPJIFRUVacKECRlbBRvJRTAPAGnS2DcDG7ZtU3VVleTzSaWlMdtxuVzyer2UhwTQqOnTp0uS9uzZo08//VTPPPOM3njjDa1cuVIPPfRQhnuHZCCYR25Zs6Z553fqJPXpk5SuAPVF+2ZgWVmZNmzb1qR2qPUOIF7FxcV1fn/rrbd04oknat68ebruuut00EEHZaZjSBomwCI3dO0qDRzY/Nu336a962jZvB5PQl91U+sdQCKGDx+uQw89VNZarVq1KmL/u+++qwsuuEA9evRQ27ZtVVBQoP/7v//TF198EXHsqlWrNHXqVB111FHq0qWL2rVrp0MOOUTXXXedtsUYpKioqNA111yjXr16qV27djr00EN13333aePGjTLGqKioqM7x69ev17Rp0zR48GDtu+++ysvLU+/evfWzn/1Mmzdvjmi/tLRUxhgVFxdrzZo1OvPMM+XxeNS+fXuddNJJWrFiRWJPXBZjZB65YcCA5rfRQFoDkCoFbrd6h43WF/r9ktcrFRZGPb6Uf6cAkqRNmzZ1fn/qqaf0s5/9THl5eTrnnHNUUFCgTz75RE888YSef/55vfPOOzrwwANrj3/88cf1zDPP6KSTTtKpp56qmpoarVq1Svfdd59efvllvfvuu+rYsWPt8VVVVTr55JO1evVqHX300br44otVUVGhWbNmadmyZVH7uGjRIj366KMaOXKkhg0bprZt2+rDDz+s7dPKlSvVs2fPiPNWrlyp2bNna+jQoZo0aZI2bdqkhQsX6pRTTtGaNWvUr1+/JD2LmUcwDwAA0EK8+eab+uijj9S2bVsNGTKkdvv69es1efJkeb1evfHGG3UC5Ndff12jRo3S1KlT9cwzz9Ruv+mmm/Twww/XWSFakp588klNmjRJ8+bN04033li7/e6779bq1at10UUX6Q9/+EPtt5K33HKLjjnmmKj9HT9+vK699lrl5eXV2f7qq6/qjDPO0B133KFHHnkk4rwXX3xR8+fPrzPS/9hjj2ny5Ml64IEHNG/evDieLWcgmAcAAC2Gk77dKozxDV1ThHLmwyfAWmt1zz33aP/996897pFHHtGePXv0wAMPRIx0n3LKKTrnnHP0/PPP67vvvqsdbe/du3fUa1566aX6xS9+oVdeeaVOMP/000+rVatWuuuuu+qkFxYUFOiaa67RrbfeGtFWtFF3SRo1apT69++vV155Jer+4cOHR6TsXHrppZoyZYree++9qOc4FcE8ACQoWWU3yyoq5MnLk7tdu6S0BwAhM2bMqPO7MUZPPvmkJk6cWGf722+/LUl644039P7770e089VXX6m6ulrr16/XoEGDJAU+IDz22GP605/+pLVr16qiokI1NTW153z++ee1P3/77bfasGGDCgoK5PV6I9o/4YQTovbfWqvf//73Kikp0T//+U9t27atzpyhtm3bRj1v8ODBEdvatGmj7t27x8zndyqCeQBogoZqxSfKWiv/rl0E8wCSzgb/v9qxY4fefvttXXbZZZo8ebJ69+6tk08+ufa4r7/+WlIgFaYh27dvr/35xz/+sZ555hn16dNH5557rnr06FGbDjNnzhzt2rWr9thvgwUmunfvHrXdWNt/8YtfaM6cOdp///01evRo9ezZU/n5+ZKkkpISlZWVRT3P4/FE3d66deucKyBAMA8ATZDMVWTD2SxYIRdoCZKRuuJEHTp00Kmnnqrnn39exxxzjCZMmKCPP/5Y7du3lyS5gxPxKyoq1KlTp0bbW7lypZ555hmdeuqpevnll9W69fchZU1NjWbPnl3n+FCbW7dujdpetO1fffWV5s6dqyOOOEIrVqyoM5lWkv74xz822s+WgGAecLqRIzPdgwYV5tg1C4K3dChdujRimy8stSeVub+NLUxVXl5OnXvAgY488khdfvnlevTRR3X//ffrlltukSQdf/zxWrVqlZYtW6Yzzzyz0XY+/fRTSdI555xTJ5CXpPfee0+VlZV1tnXq1El9+vSRz+eTz+eLSLVZvnx5xDU2btyompoajRo1KiKQ37x5szZu3NhoP1sC6swDThT8ihFIldDCVOHCK1bU3wfAOW699Vbl5eXpnnvuqc0fnzJlitq0aaNrr71W69evjzhn9+7ddcpHhoLx+oMKX331lX7+859Hve4ll1yimpoa3XTTTXW+jSwvL9ecOXMijg9dY/ny5XVSY7Zv367LL79ce/fujefh5jyCecCJiooI6JFy9fNKw0fSci3nFGhJevbsqcmTJ8vv99emwxx66KF66qmn5PP51L9/f5199tm67rrrdPXVV2vMmDHaf//9dfnll9e2ceyxx2r48OFatGiRhg0bpl/+8peaMGGCjjjiCLVv314HHHBAxHV/+ctfauDAgfrTn/6kQYMGadq0abriiis0cODA2gmwrVp9H5r26NFDF110kd577z0NHDhQ1113nSZNmqT+/ftr48aNGjhwYGqfKIcgzQZworFjAzcHCK/4UhilgkFL6UMsEX1bsyawGnFhYdR0ofBRsFTl/sZK3ykoKNCGDRtSck0A6XXTTTfp8ccf19y5c3XNNdeoe/fuGjdunI466ijde++9Wrp0qV599VV16NBBBxxwgC644AL9+Mc/rj3f5XLpueee06233qqXXnpJc+fOVc+ePTVp0iTdeuutOvzwwyOumZ+fr6VLl+q2227T3/72N91///066KCDdPPNN2vEiBF69tlnI/L1n3zySfXp00d//vOf9fDDD2vffffVOeeco5kzZ+r8889P+fPkBATzAOBQqcqZLysrk8fjqZ0QB8B5GptU3717d+3YsSNi+4ABA1RSUhLXNbp06RJz8aVYqXgej0dz587V3Llz62x//PHHJUmHHXZYne3t27fXrFmzNGvWrIi2ov0fWFhY2OBjz8UUQdJsAMBB6q+0mArWWvn9/pRfB0DL88UXX0Rs27Rpk26//Xa1bt1aZ599dgZ65WyMzANIm/KKChUw2tssXq9XPp8v5TnrlMoEkArnn3++9uzZo0GDBsnj8cjn8+mFF17Qzp07ddddd0XNtUfDCOYBpFT4Iks+v59gvpkKCgpSWhLSSUvdA3Ce8ePH63e/+50WLlyoiooK7bPPPjruuOM0ZcoUnXfeeZnuniMRzANIKa/How3B0mfJXmgp15T6fFJVleTzSRkKqn0+X22Kjc/n07JlyxqsNw8ATXHllVfqyiuvzHQ3cgrBPFDfmjXpuU6nTlKfPum5VgYVuN21wTwihX9zkY1C9eYJ5gEgOxHMAyFduwZKAqZLuj40IKt5PR75/P6sD+gBANmJYB4IGTAgfdciLxlBBW533XkEfr/k9UopqiHfmNLS0jql23w+n4wxKi8vZ3QeALIQpSkBAHUYY+r8bq3NydrMAJALCOYBAHV4PJ6IgJ5UGwDITkkL5o0xvYwxTxljvjDG7DLG+Iwxc4wxnRNo6xhjzB+MMZuDbW01xrxhjLkkWf0FAETndrvVu3dveb3eTHcFANCIpOTMG2MOlrRC0n6SFkv6SNIQSVMlnW6MGW6t/TrOtqZIekDSNkkvSvpcUhdJR0j6oaTfJqPPAAAAgNMlawLsPAUC+auttQ+GNhpj7pN0raRZkiY31ogxZpSkuZL+IekCa+139fa3SVJ/AQAAAMdrdppNcFR+lCSfpIfr7Z4uaYek8caYDnE0d7ekSkk/rR/IS5K1dk/zegsASJby8vJMdwEAWrxk5MyPDN6/aq2tCd8RDMjfktRe0vENNWKMOULSkZJelfSNMWakMeZ6Y8x1xphTjDFM1gWANAqtBuv3+1VWVqby8nK5XK46+wEAmZWMNJt+wfv1MfZ/osDIfV9JrzfQzrHB+68klUo6sd7+fxtjzrPWftpYh4wxq2LsOrSxcwHkttImBqAuY+T1eOrWgs9hLpcrauWaUHlKr9erDRs2SKLCDZCN6leiasz8+fNVVFSUms4gLZIRzIf+wlXE2B/a7mmknf2C95cpMOn1TEnLJXWXdJukcZJeNMYMsNbuTri3AFoclzEJr7Baba18fn+LCea9Xq98Pl/UQL26uloFBQW1wTyA7DN9+vSIbXPmzFFFRYWmTp0qj8dTZ9/AdK58jpTIphVgQ2k0LkkXWWvfDv7+bbAk5aGSBks6X9IfG2rIWjso2vbgiP0xyekuAKfwejzy+f3NCuhbioKCgtqVXkOrwfr9/sx2CkDciouLI7aVlJSooqJC11xzDSVnc1Ay8tBDI++xhq1C2/2NtBPavyUskJckWWutAiUvpUDJSwCIW4HbrRG9e6vQ623SDd+rqqrKdBcAJFlhYaGMMdq9e7dmzpypfv36KS8vrzbtpri4WMYYlZaWRpzr8/lkjIlI0SkqKpIxRp999pkeeughHX744WrXrp28Xq/uvPNO2eDgyF//+lcNGTJEHTp00H777acpU6aosrIy4jrGGBUWFuqLL77Q+PHjtd9++yk/P1+DBg3SH/7wh2Q/JY6UjJH5j4P3fWPsPyR4Hyunvn47/hj7twXv8+PrFgCgOcInuxLMA7nr/PPP1/vvv68zzjhDY8aM0X777df4SY24/vrrVVpaqrPPPlujRo3Sc889p1tuuUW7d+9Wly5dNG3aNI0ZM0YjRozQP/7xDz388MOqrq7WI488EtHWtm3bNGzYMHk8Hk2cOFF+v19/+ctfdPHFF+vzzz/XDTfc0Oz+OlkygvmlwftRxphW4RVtjDEdJQ2XtFPSO420844CZSy9xpgO1tod9fYfEbz/LAl9BpAhy8rKWtSEUicLn+wKIHeVlZXpP//5j7p165a0NletWqV//etf6tmzp6TAKP8PfvAD3X333Wrfvr1WrVqlww47TJK0a9cuHX300Xrqqac0Y8aMiA8T//rXv3ThhRfqT3/6k1q1CiSVTJs2TYMGDdItt9yi888/X3369Ela352m2Wk21toNCpST9Er6eb3dMyR1kPS78ODcGHOoMaZOZRlr7U5JT0pqJ+kOEzYd2xgzQFKRpL2S/tbcPgNIL1dYdYXQhFJkv1DuPJBTjHHOLU1uv/32pAbykvSrX/2qNpCXJI/Ho3POOUc7d+7UFVdcURvIS1JeXp5+/OMfa/fu3Vq3bl1EWy6XS7/5zW9qA3lJOuigg3T11Vdrz549+t3vfpfUvjtNsibAXilphaS5xphTJK2TdJwCNejXS7ql3vGhV6r+v9RfKVCS8hpJQ40xbylQzeY8BYL8a4IfHoDcsGZN7H2dOkk5MtJQfwJqS5pQCgDZbsiQ5E9HHDx4cMS2Aw44QJI0aFBknZJQ4L958+aIfQceeKAOOuigiO2FhYWaMWOGPvjgg+Z219GSEsxbazcYYwZLminpdEk/lPSlpAckzbDWbmvo/LB2vjXGjJB0k6QLJU1RYEXY5ZLusda+moz+AhnXtavUWDmwhgJ9hylwu1Xgdje5xjsAIPV69OiR9DbdUVIpW7du3ei+PXv2ROzr3r171GuE+l1REas6esuQtNKU1tpySRPjPDbmd0fW2u0KjOTXH80HcseAAQ3vj1I5AACQBHwzGCHWQlOhtJa9e/dG7EtnydqtW7dG3b5lyxZJ0T8ctCTJKE0JAACAHNO5c2dJUnl5ecS+lStXpq0fmzZtki/KN7uhkplHH3102vqSjQjmAQAAECGUSz9//vw6o/Pl5eWaOXNm2vpRXV2tG2+8UTU1tQUT9dlnn2nu3Llq3bq1xo0bl7a+ZKNsWgEWAAAAWeK4447TiSeeqDfffFNDhgzRySefrK1bt+r555/X6NGjo47Yp8KRRx6pd999V4MGDdKoUaNq68z7/X7Nnj1bBx98cFr6ka0YmQcAAEBUixcv1qRJk7R582Y9+OCD+uCDDzR79mz95je/SVsfOnfurBUrVqh///6aP3++nn76aR100EH6/e9/3+IXjJIYmQcANEO0Zd5jcblc8nq91K8H0ixavrkU3/vX4/Ho8ccf1+OPPx6xz0aZTFxSUqKSkpKobRUXF6u4uDjqvqKiIhUVFcXsxwEHHKAFCxY02t+WiGAeABC30tJSlZWV1fkjboyRx+NptKJEdXW1fD4fwTwAJBFpNgCABtUvW+fxeOpss9bGXaauuro6mV0DgBaPkXkAQIPy8vK0a9eu2t/dbnfUUfjCwsKYbTQlHQcAED+CeQBAg9q1a6d27drp4IMPjkiRIUgHkErR8vJRF8E8AMShNMYEsqSrqpJ8Pqm0NCsmjBpjav+Yku8OANmHnHkAiMEVY4nzdAlNGM0kj8dT+zP57gCQfQjmASAGr8eTFQF9JjVWoQYAWpJsTPshzQYAYihwu1WQ7mDW75e8XpWm96pAzgilhtXU1KhVK8YskVyhYL5+la9MIpgHkBHRctBdxsjr8aQ/gAaQM/Ly8lRVVaUdO3aoY8eOme4OcsyOHTskBf6dZQuCeSCbrVkTua1TJ6lPn7R3JRlcxqi6ga8oq62Vz+8nmAeQsI4dO6qqqkpbtmyRJHXo0EHGmKwaSYWzWGtlrdWOHTtq/11l0wdFgnkgG3XtKg0cGH1ftADfIbwej3x+f6MBPQAkqkuXLtqxY4d27typzZs3Z7o7yEHt27dXly5dMt2NWgTzQDYaMCD6dofX9G4oBz1tpR8B5LRWrVqpoKBA33zzjb777jvt2rUrKyctwlmMMcrLy1PHjh3VpUuXrJqPQTAPAABySqtWrdStWzd169Yt010BUi57PlYAAAAAaBKCeQAAAMChCOYBAElRXl6e6S4AQItDMA8ASJjL5ar92RfnJOZly5YR+ANAkhDMAwAS5vV6a3+urq6OeVx40F9dXR134A8AaBjBPAAgLhUVFRHbCgoK4jrX6/VGBPQAgOYjmAcAxBQegPv9/oTbKSgo0IgRI5LQIwBAOIJ5AEBM4Wk0LLwDANmHYB4AEFO8aTQAgMwgmAcAAAAcimAeAAAAcCiCeQAAAMChCOYBJ1qzRtq4MdO9AAAAGdY60x0A0ARdu0oDBwZ+XrMmkz0BAABZgGAecJIBAwL3paUZ7UaqlTawOqjLGHk9HhW43enrEAAAWYo0GwBZwWVMXMdVWytfMxYvAgAglxDMA8gKXo+nSQE9AAAgzQZAlihwuxtNnWko/QbpUV5ezkJSAJBFCOYBAA0yxsgGvw3x+XwNBvOlwfkcLpdLXq+XwB8AUoxgHgDQII/Ho23btkmSqqurI/a7XK6I7dXV1XEH/o3hgwEAxEbOPACgQe5G0p+8Xq9cLlfE9liBf1OFPhgAACIxMg8AaJaCgoI6o+YNjbh7vV75fL6ogX5Dmno8ALQUBPMAgLSpH/g3Jt5UHABoqUizAQAAAByKYB4AAABwKIJ5AAAAwKEI5gEAAACHIpgHAAAAHIpqNgCQ5dJd0YVFmgDAORiZB4AslMjiSsnCIk0A4BwE8wCQhWKtqpouLNIEAM5Amg0AZKGmLq6ULCzSBADOQjAPwJFKsywNxGWMvB6PCtzuTHcFANCCEMwDcAyXMaq2NtPdiKraWvn8foL5epo70h+eu99QW0zaBdBSkTMPwDG8Ho9cxmS6GzFl6weNdMtErj+TdgG0VIzMA3CMArc7K0e+sy3lJ9O8Xq98Pl/aJ9EyaRdAS0QwDwBIqmRO3g1PrTn44IOjtsukXQAtGWk2AICsFZ6yQxoNAEQimAcAZC2v11v7M2k0ABCJYB4AkLWoTgMADSOYBwAAAByKYB4AAABwKIJ5AAAAwKEoTQk42Zo1me4BJKmq6vuf/f6MdQMA0PIQzANO1LWrNHBgpnuBkPCSiWHVVxLWtWvz20ih8vJyJqYCQJYgmAecaMCATPcA4cIXLSoszFQvUsoYI2utpEC9d4J5AMgO5MwDABrl8Xhqf6beOwBkD4J5AECj3G53prsAAIiCYB4AAABwKHLmAQA5ozR8/kI9LpdLXq+XfH8AOYVgHgCSqKFg0gl8YZV5Qo8ltM2fpWU3XS5XXHn81dXVTN4FkHMI5gGgmeINJnNBKLBP54eW8A8Yy5Ytixhd93q98vl8cQf0AJBLCOYBoJmaEkw6UXhZykz75ptvJKlOMF9QUNDoaLvTvzEBgFgI5gGgmeIJJp0iPOgtDNbMLy8vl8/n04YNGzLSp/APE36/n8o6ABCGYB4AEFUosA9NHA3/5qEwjYtjlZeXa8OGDfL5fFnzDQEAZIuklaY0xvQyxjxljPnCGLPLGOMzxswxxnRuRpsnGmOqjTHWGHNHsvoKAIjO5XJFbAtNHM2UXPnWAwBSISnBvDHmYEmrJE2U9J6k+yVtlDRV0tvGmK4JtNlR0tOSdiajjwCAxnm93pgBPQAg+yQrzWaepP0kXW2tfTC00Rhzn6RrJc2SNLmJbT4gyS3pruD5AIAUq5//z8RRAMhuzR6ZD47Kj5Lkk/Rwvd3TJe2QNN4Y06EJbZ6rwCj/1ZK+aG4fAQDJVV5enukuAACUnDSbkcH7V621NeE7rLXfSXpLUntJx8fTmDFmP0mPS3rWWrsgCf0DACRBePpNJnPoAQDfS0aaTb/g/foY+z9RYOS+r6TX42jvcQU+ZDQ1LaeWMWZVjF2HJtomALR0Xq+3tjwlOfQAkB2SMTIfKvhbEWN/aLunsYaMMZdKOkfSldbarc3vGgAgWagqAwDZJ2vqzBtjvJLmSPqrtfYvzWnLWjsoxjVWSTqmOW0DAAAA2SIZI/OhkfdYS/KFtvsbaecpSZWSrkxCnwAAAICcl4yR+Y+D931j7D8keB8rpz7kGAUC//8aY6Ltv8UYc4ukxdbaMU3tJAAAUubKbYZW0iVdCUAyJSOYXxq8H2WMaRVe0Sa48NNwBRZ+eqeRdn6rQNWb+g6RdKKkNQosTPVBczsMAGhZXC5XxifthlbSJZgHkEzNDuattRuMMa8qULHm55IeDNs9Q1IHSY9Za3eENhpjDg2e+1FYO1dHa98YU6RAMP+itfbW5vYXAOB8y5Yta9Iot9frlc/ny4qAHgCSKVkTYK+UtELSXGPMKZLWSTpOgRr06yXdUu/4dcH7qPk0AADUZ4yRtVZS00e5669sm26spAsgVZIxAVbW2g2SBksqUSCIv07SwZIekHS8tfbrZFwHANByeTwehc+pYpQbAJJYmtJaWy5pYpzHxj0ib60tUeBDAgCgBXO73XK73XVWn413xJvJpwByVVJG5gEASCWXy9Ws80NpOQCQawjmAQBZz+v1JiWgB4BckzUrwAIAEEv4BNbw1JrCwsJGz2XyKYBcxsg8AAAA4FAE8wAAAIBDEcwDAAAADkUwDwAAADgUwTwAAADgUATzAAAAgENRmhIAgDSiVGYkVugFEsfIPAAAKdbcBa9yHSv0AoljZB4A4FjxjHKHB4mZGhWvqKiQ3++XtTZt1zTGyOPxyO12p+2azcEKvUBiCOYBAI7icrkcF/i53e6MBNUul0sjRoxI+3WbgrQjoHlIswEAOIrX6yVtJU5O+9ADoOkYmQcAOEpBQUGTJkqGj/wWFhYmv0NZiNFuoOUgmAcAJKS8vNxx1UeyJcileguAZCHNBgAQt/D0FqdUH8nGlByqtwBIFoJ5AEDcvF5v7c9OycfO1hx7pzx/ALIbaTYAgLgVFBRow4YNme5GkzQ1xz7VsiXVB0BuYGQeAAAAcCiCeQAAAMChCOYBAAAAhyKYBwAAAByKYB4AAABwKIJ5AAAAwKEoTQkAALJCrLKdrJgLxMbIPAAAyJh4FvRixVwgNoJ5AAByWHl5eaa70KB4V+hlxVwgOtJsAAAJYzXT7EwBcblctcGvz+fLqr7V19gKvfwbAxrGyDwAoEniGUVtSbIxBcTr9db+zIg2kNsI5gEATRJvWkRLkm0BczaPxANILtJsAABN0lhaREtCCgiATCOYBwAgQ5LxYSAbc/YBpA9pNgAApFGyU5SyMWcfQPoQzAMAkEapmHOQbTn7ANKHNBsAANIomXMOyNkHwMg8AAAA4FAE8wAAAIBDEcwDAAAADkUwDwAAADgUE2ABAIAjZGLCL3X8ke0YmQcAAFkr2WU8m4o6/sh2jMwDAJAEzRk1TvXob7pGtFPxOLxer3w+X0Zr6VPHH9mMYB4AgAS5XK6kBHqh0d9kBsHJ6ltTpOJxJLMuf1NRxx9OQJoNAAAJSuZqrskOvFOx0mw8GMUG0ouReQAAEpSMUeNUjf6me0SbUWwgMxiZBwAAAByKYB4AAABwKNJsAADIAZlOcwkv30g9eCB9GJkHAMChMl2DPZtQDx4tFcE8AAAOlamKNdmKSjpoiUizAQDAoTJZg72+8NSawsLCjF0baGkYmQcAAAAcimAeAAAAcCiCeQAAAMChCOYBAAAAh2ICLAAAWaL+RE5qpwNoDME8AAAZ5HK5YpZUDNVOd1own+7qMqlasCrTC2EhMS3tQzDBPAAAGeT1euXz+RoM6J2goQ8luSBZC1IZY+TxeOR2u5PSHiI59UNwogjmAQDIoFi14p02EtzYhxInMsbIWpvUNq218vv9BPMplkv/DhtDMA8AAJotkwtYpWrBqvLy8pR9QEn3wlothdM+BCcDwTwAAEAUyf6A0hIDTaQepSkBAAAAhyKYBwAAAByKYB4AAABwKIJ5AAAAwKEI5gEAAACHIpgHAAAAHIrSlAAAIGdkc/nH8FVkm9JPl8slr9fbYlY0RdMwMg8AABzN5XJlugspVV1dXeeDABCOYB4AADia1+ttEQE9EA1pNgAAZDlSMhqW7JVaUyX8dSwsLGzyOUA0SRuZN8b0MsY8ZYz5whizyxjjM8bMMcZ0jvP8DsaYi40xfzDGfGSM2WGM+c4Ys9IYc50xpm2y+goAQLZLdKSZlAygZUlKMG+MOVjSKkkTJb0n6X5JGyVNlfS2MaZrHM2MkLRA0mhJ/5H0oKQ/SOop6R5JS40x7ZLRXwAAsl1zUkdIyQBajmSl2cyTtJ+kq621D4Y2GmPuk3StpFmSJjfSxhZJ4yT91Vq7O6yN6yWVShom6eeS7k1SnwEATVBaWhp3agCaL5HUEVIygJan2SPzwVH5UZJ8kh6ut3u6pB2SxhtjOjTUjrV2jbX29+GBfHD7d/o+gC9sbn8BAPHL9UmFAHJXaWmpli1bpvLy8kx3JaWSkWYzMnj/qrW2JnxHMBB/S1J7Scc34xp7gvd7m9EGAKCJWkKVEAC5o/7/Vy1hDkky0mz6Be/Xx9j/iQIj930lvZ7gNS4N3v89noONMati7Do0wesDQIvklCohACAFBiB8Pl+deSO5PockGcG8O3hfEWN/aLsnkcaNMVMknS5pjaSnEmkDAAAAuS98AKKlzCHJ6jrzxpjzJM1RYHLs+dbaPQ2fEWCtHRSjvVWSjklaBwEAABIQHmi2xLUBkDzJyJkPjby7Y+wPbfc3pVFjzBhJf5L0laRCa+3GRDoHAACQDWLNP2kJed1InWQE8x8H7/vG2H9I8D5WTn0EY8yFkv4qaaukk6y1HzdyCgAAQFZraEJ5rud1I3WSkWazNHg/yhjTKryijTGmo6ThknZKeieexowxF0t6WtLnkkYyIg8AAHJBtAnlLSWvG6nT7JF5a+0GSa9K8iqwqFO4GZI6SPqdtXZHaKMx5lBjTERlGWPMBEm/lbRJ0okE8gAAAEBsyZoAe6WkFZLmGmNOkbRO0nEK1KBfL+mWesevC96b0AZjzEgFqtW0UmC0f6Ixpt5p8ltr5ySpzwAAAICjJSWYt9ZuMMYMljRTgTKSP5T0paQHJM2w1m6Lo5ne+v6bgktjHFOmQHUbAAAAoMVLWmlKa225pIlxHhsx5G6tLZFUkqz+AAAAALkuq+vMAwCAxJSWllK/HGgBklGaEgAAZIH6ZQ+pXw7kPoJ5AAByRLQ65tQvB3IbaTYAAOSI8Drm1C8HWgZG5gEAAACHYmQeAADAAfi2pWl8Pp+MMfJ4PJnuSkoxMg8AAJCl6s+BQNNYa+X3+1VeXp7prqQMwTwAAECWijapGfExJrCskbU2p6s6kWYDAACQpcInNaNpFi9erG3btknK7apOjMwDAAAg57jd7kx3IS0I5gEAAACHIpgHAAAAHIpgHgAAAHAognkAAADAoahmAwAAkAVydVEol8slr9dLVZ4UIZgHACDHZUOQSEAXncvlyumyiVKgLKTP5+O1TxHSbAAAyEHZttBQKKBDXS1lUahc/8CSSYzMAwCQg7xer3w+X1YFUdnUl2yR64tCZcO3QrmOYB4AgByUTUEiAR2QOqTZAAAAAA5FMA8AAAA4FME8AAAA4FAE8wAAAIBDEcwDAAAADkUwDwAAADgUwTwAAADgUATzAAAAgEOxaBQAAAByktfrlSQVFhZmtB+pxMg8AAAA4FAE8wAAAMhppaWlme5CyhDMAwAAIOe4XK5MdyEtCOYBAACQc7xeb4sI6JkACwAAgJxTUFCggoKCTHcj5RiZBwAAAByKYB4AAABwKNJsAAAAkHKZriiTq7XmCeYBAEDaZDqgQ3qVlZXJWpvpbuQ00mwAAEBKtYSKIojO4/HIGJPpbuQ0RuYBAEBKeb1e+Xw+VVdXZ7orSDO32y23253pbuQ0gnkAAJBSLaVEIJAJpNkAAAAADkUwDwAAADgUwTwAAADgUATzAAAAgEMRzAMAAAAORTAPAAAAOBTBPAAAAOBQBPMAAACAQxHMAwAAAA5FMA8AAAA4FME8AAAA4FAE8wAAAIBDEcwDAAAADkUwDwAAADgUwTwAAADgUATzAAAAgEMZa22m+5A2xpiv8/Pzuxx22GGZ7goAAABy2Lp161RZWfmNtbZrKq/T0oL5zyR1kuTLwOUPDd5/lIFro3G8PtmN1ye78fpkN16f7MdrlN0SfX28kr611h6U3O7U1aKC+UwyxqySJGvtoEz3BZF4fbIbr0924/XJbrw+2Y/XKLtl++tDzjwAAADgUATzAAAAgEMRzAMAAAAORTAPAAAAOBTBPAAAAOBQVLMBAAAAHIqReQAAAMChCOYBAAAAhyKYBwAAAByKYB4AAABwKIJ5AAAAwKEI5gEAAACHIpgHAAAAHIpgPsWMMb2MMU8ZY74wxuwyxviMMXOMMZ0z3bdcEnxebYzblhjnDDPGvGSM+cYYU2mM+Zcx5hpjjKuB65xljCk1xlQYY7YbY941xkxI3SNzFmPMBcaYB40xy4wx3waf/wWNnJOW18EYM8EY817w+Irg+Wcl+lidqCmvjzHG28B7yhpj/tTAdZr0XBtjXMaYa4OvfWXw38JLxphhyXjcTmCM6WqMmWSMecYY82nweagwxiw3xlxmjIn695r3T3o09fXh/ZMZxpjfGGNeN8aUhz0XHxhjphtjusY4x/HvIRaNSiFjzMGSVkjaT9JiSR9JGiJppKSPJQ231n6duR7mDmOMT5JH0pwou7dba++pd/y5khZKqpL0Z0nfSDpbUj9Jf7PWXhjlGlMkPSjp6+A5uyVdIKmXpHuttdcn59E4lzFmjaSjJG2XtFnSoZJ+b60dF+P4tLwOxph7JF0X7NPfJLWVdJGkLpKustY+lPCDdpCmvD7GGK+kzyT9U9KzUZr7j7X2b1HOa9JzbYwxkv6iwGv4saTng8f+WFI7Sedbaxc3+cE6jDFmsqRHJH0paamkTZK6SzpPkluB98mFNuyPNu+f9Gnq68P7JzOMMbslrZa0VtJXkjpIOl7SYElfSDreWlsednxuvIestdxSdJP0iiQbfKHCt98X3P5opvuYKzdJPkm+OI/tpMCbfJekwWHb2ynw4ctKuqjeOV4F3uxfS/KGbe8s6dPgOUMz/Txk+qbAB9VDJBlJhcHnZUEmXwdJw4LbP5XUuV5bXwfb8zbncTvl1sTXxxvcX9KE9pv8XEv6SfCctyS1C9t+bPDfxleSOmb6uUvDa3OyAkFEq3rbeygQOFoFArPQdt4/2f368P7JzOvULsb2WcHnaV7Ytpx5D5FmkyLBUflRCgSZD9fbPV3SDknjjTEd0tw1BD5B7yvpT9balaGN1toqSbcGf72i3jmXSsqT9JC11hd2zjZJdwZ/nZyqDjuFtXaptfYTG/yfqhHpeh1Cv88KHhc6x6fAezNP0sQ4+ut4TXx9EpHIcx16jW8Nvvahc95XYNRrXwX+reQ0a+0Sa+3z1tqaetu3SHo0+Gth2C7eP2mUwOuTCN4/zRT+HNTzl+D9IWHbcuY9RDCfOiOD969GefN/p8Cn6PYKfP2D5MgzxowzxtxsjJlqjBkZI+ft5OD936Pse1PSTknDjDF5cZ7zcr1jEJ90vQ68ds1zgDHm/4Lvq/8zxhzZwLFNeq6NMe0UGLXaKWlZPOe0UHuC93vDtvH+yR7RXp8Q3j/Z4ezg/b/CtuXMe6h1c05Gg/oF79fH2P+JAiP3fSW9npYe5b4ekn5Xb9tnxpiJ1to3wrbFfG2stXuNMZ9J6i+pj6R1cZzzpTFmh6Rexpj21tqdzXkQLUjKX4fgN189FZg38WWUPnwSvO/bjMeR604L3moZY0olTbDWbgrblshzfbAkl6SN1tpogVCLf32MMa0lXRL8NTwY4P2TBRp4fUJ4/2SAMeZ6SfsoMJ9hsKQTFAjkfx12WM68hxiZTx138L4ixv7Qdk/qu9IizJd0igIBfQdJAyQ9pkBO2svGmKPCjk3ktYn3HHeM/YiUjteB92Hidkq6XdIgBfJBO0s6SYHJf4WSXq+XJpjK19MTY39L8GtJR0h6yVr7Sth23j/ZIdbrw/sns65XIKX5GgUC+b9LGmWt/W/YMTnzHiKYR06w1s4I5jRutdbutNb+x1o7WYHJxvmSijPbQ8BZrLVfWWtvs9auttb6g7c3FfhG8V1JP5A0KbO9zG3GmKsVqIDxkaTxGe4O6mno9eH9k1nW2h7WWqPAAN95Coyuf2CMOSazPUsNgvnUaWykNrTdn/qutGihiUknhm1L5LWJ95xYn74RKR2vA+/DJAt+nf9E8Nd0va/8MfbnrGD5uwcUKLE30lr7Tb1DeP9kUByvT1S8f9IrOMD3jAIforpK+m3Y7px5DxHMp87HwftYeVChGdWxcuqRHKGv1MK/zoz52gTzHw9SYCLTxjjP2T/Y/mby5Zsk5a+DtXaHpM8l7RPcXx/vw8REvK8SfK43SKqW1Cf4msdzTs4zxlyjQB3r/ygQKEZb+I73T4bE+fo0hPdPmllryxT44NXfGNMtuDln3kME86mzNHg/ykSuDNdR0nAFcureSXfHWphQtaDwN+OS4P3pUY4/UYEqQyustbviPOeMescgPul6HXjtki/a+0pq4nMdLAG3QoHXekQ85+Q6Y8yNku6XtEaBQPGrGIfy/smAJrw+DeH9kxkHBO+rg/e58x6qX3ieW1IXL2DRqPQ8z4dJ6hBlu1eBmeJW0s1h2zspMDLSlIUiDhKLRjX1dSlU44tGpfx1EIveJPr6HKN6C+QEt58SfM6spGHNfa4V36I3nTL9fKXpNflV8LlYKalLI8fy/snu14f3T/pfn76S3FG2t9L3i0a9FbY9Z95DJtggUiC4cNQKSftJWqxAaaPjFKhBv16BN/LXmethbjDGFCswCelNSWWSvlOgZNeZCrwpX5L0I2vt7rBzxiiwpHKVpD8psITzOQou4SxprK335jDGXCVprpqwhHNLE3xexwR/7SFptAKjT6EayP8Lf57S9ToYY+6V9AvVXUr7xwrkULaI5eilpr0+wfJ5hyjwf9jm4P4j9X095F9Za++Ico0mPdf1lqP/SIHl6LuqhS1Hb4yZIKlEgVHDBxV9/o3PWlsSds4Y8f5Ji6a+Prx/0i+Y/nSXpOWSPlPg33h3BaoI9ZG0RdIp1tq1YeeMUS68hzL9SSrXb5IKFCib+GXwBS+TNEdhn864Nfs5PknSHxX4j8yvwAIe/5X0DwXq/5oY5w1XINDfJqlS0r8lXSvJ1cC1zpb0hgIfGHZIel+BesEZfx6y4aZA1SDbwM2XqddBUlHwuB3B896QdFamn7NsfX0kXSbpBQVWsd6uwOjVJgX+eI1I5nOtwJon1wZf+8rgv4WXVG/kMpdvcbw2VlJplPN4/2Th68P7JyOv0RGSHlIgBep/CuS7VwSfy2LF+DYlF95DjMwDAAAADsUEWAAAAMChCOYBAAAAhyKYBwAAAByKYB4AAABwKIJ5AAAAwKEI5gEAAACHIpgHAAAAHIpgHgAAAHAognkAAADAoQjmAQAAAIcimAcAAAAcimAeABzAGFNojLHGmOJM9yWcMabUGGPrbcvKvgJALiKYB4AsYYzxBoPgkkz3BQDgDK0z3QEAQFzek3SYpP9luiNxcFJfAcDRCOYBwAGstTslfZTpfsTDSX0FAKcjzQYAskAwv/yz4K8Tguk2oVtRrDz0UM66MaaNMeY2Y8wGY0yVMeZjY8zlYcdNNsb82xhTaYzZbIyZYYyJ+jfAGHOcMeZvxpgtxpjdxphyY8xjxpgD4nwsjfW1tTHmZmPMJ8aYXcH2f2OMaRujvUONMSXB43YbY7YaY/5gjOkXT38AIJcxMg8A2aFUkkfSVEn/lPRs2L41wX0N+ZOk4yS9JGmPpAsk/T9jzB5JR0qaIOkFSa9LOkfSbZJ2SvpNeCPGmEsl/T9JuyQ9J6lc0iGSJkk62xhzvLV2UyIPMMwfJI2Q9LKkbyX9UNIvJe0naWK9/pwuaZGkNpKel/SppF6SzpN0pjFmpLV2dTP7AwCORTAPAFnAWltqjPEpEMyvsdYWh+83xhQ20sSBko6w1vqDx9+rQKrL/ZL8ko601n4e3FesQFB8vTHmXmvt3uD2vpIeleSTdFLo+OC+UyS9KukBST9K9HEGHSypv7X2m2DbtyjwAeYSY8xN1totwe2dJf1RgQ8dJ1pr14b15whJ70h6QtIxzewPADgWaTYAkBumhQJ5SbLWbpS0XIER/dvDA/Pgcc9L6iapZ1gbVygwAj41/PjgOa8rMFJ/tjGmYzP7emMokA+2vUPS7xX4mzQ47LhLgv2fHh7IB8/5j6THJR1tjDm8mf0BAMdiZB4AcsPKKNu+CN6virIvFKz3klQW/Hlo8P4kY8yxUc7ZT5JLUt8YbcYrWl/Lg/edw7aF+nNUjJr1fYP3h0laG2U/AOQ8gnkAyAHW2ooom/cG7xva1yZsW9fg/Q2NXG6fJnQtQvg3CFH644rSn8vVsGb1BwCcjGAeABASCvrd1tpvM9qTgFB/jrLW/iujPQGALEXOPABkj+rgvavBo1LnneD9iAxdv75s6w8AZB2CeQDIHtskWQUq02TCQwqUtbw/WNmmDmNMW2NMOgPr+QpU4plujBkSpT+t4qjyAwA5jTQbAMgS1trtxph3JY0wxvxe0noFRuufS9P1PwrWmX9K0ofGmL8H+9BGgQ8YIyT9V9KhaerP18aYCyQ9I+kdY8zrkj5U4ANPgQITZLtKapeO/gBANiKYB4DsMl6B2vCnS/qJJCNpswK131POWrvAGPNPSddJGilplKQdClTG+ZukP6ejH2H9ed0Yc6Sk6yWNVuADxe5gf5ZIWpjO/gBAtjHW2kz3AQAAAEACyJkHAAAAHIpgHgAAAHAognkAAADAoQjmAQAAAIcimAcAAAAcimAeAAAAcCiCeQAAAMChCOYBAAAAhyKYBwAAAByKYB4AAABwKIJ5AAAAwKEI5gEAAACHIpgHAAAAHIpgHgAAAHAognkAAADAoQjmAQAAAIcimAcAAAAc6v8DKMtaMlqZGYQAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 261,
       "width": 377
      },
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "ax = plt.subplot()\n",
    "\n",
    "for name, df_ in raw_df[['president','event', 'T']].groupby('president'):\n",
    "    kmf.fit(df_['T'], df_['event'], label=name)\n",
    "    if name == 'Trump':\n",
    "        ax = kmf.plot(ax=ax, color='r')\n",
    "    else:\n",
    "        ax = kmf.plot(ax=ax, color='grey', alpha=0.5, ci_show=False)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 261,
       "width": 372
      },
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "raw_df[['president','event', 'T']]\n",
    "\n",
    "naf = NelsonAalenFitter()\n",
    "\n",
    "ax = plt.subplot()\n",
    "\n",
    "for name, df_ in raw_df[['president','event', 'T']].groupby('president'):\n",
    "    if name in ['Trump', 'Carter']:\n",
    "        naf.fit(df_['T'], df_['event'], label=name)\n",
    "        ax = naf.plot(ax=ax)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "raw_df['year'] = raw_df['start'].apply(lambda d: int(d.year))\n",
    "\n",
    "regression_df = raw_df[['president', 'T', 'event', 'year']]\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "ename": "ConvergenceError",
     "evalue": "Convergence halted due to matrix inversion problems. Suspicion is high collinearity. Please see the following tips in the lifelines documentation: https://lifelines.readthedocs.io/en/latest/Examples.html#problems-with-convergence-in-the-cox-proportional-hazard-modelMatrix is singular.",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mLinAlgError\u001b[0m                               Traceback (most recent call last)",
      "\u001b[0;32m~/code/lifelines/lifelines/fitters/coxph_fitter.py\u001b[0m in \u001b[0;36m_newton_rhapson_for_efron_model\u001b[0;34m(self, X, T, E, weights, entries, initial_point, step_size, precision, show_progress, max_steps)\u001b[0m\n\u001b[1;32m    988\u001b[0m             \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 989\u001b[0;31m                 \u001b[0minv_h_dot_g_T\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mspsolve\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m-\u001b[0m\u001b[0mh\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mg\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0massume_a\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m\"pos\"\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcheck_finite\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mFalse\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    990\u001b[0m             \u001b[0;32mexcept\u001b[0m \u001b[0;34m(\u001b[0m\u001b[0mValueError\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mLinAlgError\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mas\u001b[0m \u001b[0me\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/venvs/data/lib/python3.7/site-packages/scipy/linalg/basic.py\u001b[0m in \u001b[0;36msolve\u001b[0;34m(a, b, sym_pos, lower, overwrite_a, overwrite_b, debug, check_finite, assume_a, transposed)\u001b[0m\n\u001b[1;32m    247\u001b[0m                            overwrite_b=overwrite_b)\n\u001b[0;32m--> 248\u001b[0;31m         \u001b[0m_solve_check\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mn\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0minfo\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    249\u001b[0m         \u001b[0mrcond\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0minfo\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mpocon\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mlu\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0manorm\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/venvs/data/lib/python3.7/site-packages/scipy/linalg/basic.py\u001b[0m in \u001b[0;36m_solve_check\u001b[0;34m(n, info, lamch, rcond)\u001b[0m\n\u001b[1;32m     28\u001b[0m     \u001b[0;32melif\u001b[0m \u001b[0;36m0\u001b[0m \u001b[0;34m<\u001b[0m \u001b[0minfo\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 29\u001b[0;31m         \u001b[0;32mraise\u001b[0m \u001b[0mLinAlgError\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m'Matrix is singular.'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     30\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;31mLinAlgError\u001b[0m: Matrix is singular.",
      "\nDuring handling of the above exception, another exception occurred:\n",
      "\u001b[0;31mConvergenceError\u001b[0m                          Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-15-f5c8742b8449>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[1;32m      1\u001b[0m \u001b[0mcph\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mCoxPHFitter\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 2\u001b[0;31m \u001b[0mcph\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mfit\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mregression_df\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m'T'\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m'event'\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mformula\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m\"president + bs(year, df=3)\"\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m      3\u001b[0m \u001b[0mcph\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mprint_summary\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;36m3\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/code/lifelines/lifelines/utils/__init__.py\u001b[0m in \u001b[0;36mf\u001b[0;34m(model, *args, **kwargs)\u001b[0m\n\u001b[1;32m     52\u001b[0m         \u001b[0;32mdef\u001b[0m \u001b[0mf\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mmodel\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     53\u001b[0m             \u001b[0mcls\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mset_censoring_type\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mmodel\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcls\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mRIGHT\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 54\u001b[0;31m             \u001b[0;32mreturn\u001b[0m \u001b[0mfunction\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mmodel\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     55\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     56\u001b[0m         \u001b[0;32mreturn\u001b[0m \u001b[0mf\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/code/lifelines/lifelines/fitters/coxph_fitter.py\u001b[0m in \u001b[0;36mfit\u001b[0;34m(self, df, duration_col, event_col, show_progress, initial_point, strata, step_size, weights_col, cluster_col, robust, batch_mode, timeline, formula, entry_col)\u001b[0m\n\u001b[1;32m    284\u001b[0m             \u001b[0mtimeline\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mtimeline\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    285\u001b[0m             \u001b[0mformula\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mformula\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 286\u001b[0;31m             \u001b[0mentry_col\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mentry_col\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    287\u001b[0m         )\n\u001b[1;32m    288\u001b[0m         \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/code/lifelines/lifelines/fitters/coxph_fitter.py\u001b[0m in \u001b[0;36m_fit_model\u001b[0;34m(self, *args, **kwargs)\u001b[0m\n\u001b[1;32m    303\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0m_fit_model\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    304\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mbaseline_estimation_method\u001b[0m \u001b[0;34m==\u001b[0m \u001b[0;34m\"breslow\"\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 305\u001b[0;31m             \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_fit_model_breslow\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    306\u001b[0m         \u001b[0;32melif\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mbaseline_estimation_method\u001b[0m \u001b[0;34m==\u001b[0m \u001b[0;34m\"spline\"\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    307\u001b[0m             \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_fit_model_spline\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/code/lifelines/lifelines/fitters/coxph_fitter.py\u001b[0m in \u001b[0;36m_fit_model_breslow\u001b[0;34m(self, *args, **kwargs)\u001b[0m\n\u001b[1;32m    313\u001b[0m             \u001b[0mpenalizer\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mpenalizer\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0ml1_ratio\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0ml1_ratio\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mstrata\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mstrata\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0malpha\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0malpha\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mlabel\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_label\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    314\u001b[0m         )\n\u001b[0;32m--> 315\u001b[0;31m         \u001b[0mmodel\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mfit\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    316\u001b[0m         \u001b[0;32mreturn\u001b[0m \u001b[0mmodel\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    317\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/code/lifelines/lifelines/utils/__init__.py\u001b[0m in \u001b[0;36mf\u001b[0;34m(model, *args, **kwargs)\u001b[0m\n\u001b[1;32m     52\u001b[0m         \u001b[0;32mdef\u001b[0m \u001b[0mf\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mmodel\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     53\u001b[0m             \u001b[0mcls\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mset_censoring_type\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mmodel\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcls\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mRIGHT\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 54\u001b[0;31m             \u001b[0;32mreturn\u001b[0m \u001b[0mfunction\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mmodel\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     55\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     56\u001b[0m         \u001b[0;32mreturn\u001b[0m \u001b[0mf\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/code/lifelines/lifelines/fitters/coxph_fitter.py\u001b[0m in \u001b[0;36mfit\u001b[0;34m(self, df, duration_col, event_col, show_progress, initial_point, strata, step_size, weights_col, cluster_col, robust, batch_mode, timeline, formula, entry_col)\u001b[0m\n\u001b[1;32m    726\u001b[0m             \u001b[0minitial_point\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0minitial_point\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    727\u001b[0m             \u001b[0mshow_progress\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mshow_progress\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 728\u001b[0;31m             \u001b[0mstep_size\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mstep_size\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    729\u001b[0m         )\n\u001b[1;32m    730\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/code/lifelines/lifelines/fitters/coxph_fitter.py\u001b[0m in \u001b[0;36m_fit_model\u001b[0;34m(self, X, T, E, weights, entries, initial_point, step_size, show_progress)\u001b[0m\n\u001b[1;32m    845\u001b[0m     ):\n\u001b[1;32m    846\u001b[0m         beta_, ll_, hessian_ = self._newton_rhapson_for_efron_model(\n\u001b[0;32m--> 847\u001b[0;31m             \u001b[0mX\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mT\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mE\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mweights\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mentries\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0minitial_point\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0minitial_point\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mstep_size\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mstep_size\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mshow_progress\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mshow_progress\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    848\u001b[0m         )\n\u001b[1;32m    849\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/code/lifelines/lifelines/fitters/coxph_fitter.py\u001b[0m in \u001b[0;36m_newton_rhapson_for_efron_model\u001b[0;34m(self, X, T, E, weights, entries, initial_point, step_size, precision, show_progress, max_steps)\u001b[0m\n\u001b[1;32m   1000\u001b[0m                             \u001b[0mCONVERGENCE_DOCS\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1001\u001b[0m                         ),\n\u001b[0;32m-> 1002\u001b[0;31m                         \u001b[0me\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   1003\u001b[0m                     )\n\u001b[1;32m   1004\u001b[0m                 \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;31mConvergenceError\u001b[0m: Convergence halted due to matrix inversion problems. Suspicion is high collinearity. Please see the following tips in the lifelines documentation: https://lifelines.readthedocs.io/en/latest/Examples.html#problems-with-convergence-in-the-cox-proportional-hazard-modelMatrix is singular."
     ]
    }
   ],
   "source": [
    "cph = CoxPHFitter()\n",
    "cph.fit(regression_df, 'T', 'event', formula=\"president + bs(year, df=3)\")\n",
    "cph.print_summary(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "cph.check_assumptions(regression_df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "ename": "AttributeError",
     "evalue": "Must call `fit` first.",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mAttributeError\u001b[0m                            Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-13-9b2470913cd3>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0mcph\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mplot_covariate_groups\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\"year\"\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mvalues\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mlinspace\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;36m1977\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;36m2016\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;36m5\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0my\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m\"cumulative_hazard\"\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
      "\u001b[0;32m~/code/lifelines/lifelines/fitters/coxph_fitter.py\u001b[0m in \u001b[0;36m__getattr__\u001b[0;34m(self, attr)\u001b[0m\n\u001b[1;32m    292\u001b[0m             \u001b[0;32mraise\u001b[0m \u001b[0mAttributeError\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\"Must call `fit` first.\"\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    293\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 294\u001b[0;31m         \u001b[0;32mif\u001b[0m \u001b[0mhasattr\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_model\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mattr\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    295\u001b[0m             \u001b[0;32mreturn\u001b[0m \u001b[0mgetattr\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_model\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mattr\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    296\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/code/lifelines/lifelines/fitters/coxph_fitter.py\u001b[0m in \u001b[0;36m__getattr__\u001b[0;34m(self, attr)\u001b[0m\n\u001b[1;32m    290\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0m__getattr__\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mattr\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    291\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0mattr\u001b[0m \u001b[0;34m==\u001b[0m \u001b[0;34m\"_model\"\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 292\u001b[0;31m             \u001b[0;32mraise\u001b[0m \u001b[0mAttributeError\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\"Must call `fit` first.\"\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    293\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    294\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0mhasattr\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_model\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mattr\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;31mAttributeError\u001b[0m: Must call `fit` first."
     ]
    }
   ],
   "source": [
    "cph.plot_covariate_groups(\"year\", values=np.linspace(1977, 2016, 5), y=\"cumulative_hazard\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<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",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>model</th>\n",
       "      <td>lifelines.WeibullAFTFitter</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>duration col</th>\n",
       "      <td>'T'</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>event col</th>\n",
       "      <td>'event'</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>number of observations</th>\n",
       "      <td>287</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>number of events observed</th>\n",
       "      <td>158</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>log-likelihood</th>\n",
       "      <td>-1331.256</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>time fit was run</th>\n",
       "      <td>2020-07-12 23:59:57 UTC</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div><table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th>coef</th>\n",
       "      <th>exp(coef)</th>\n",
       "      <th>se(coef)</th>\n",
       "      <th>coef lower 95%</th>\n",
       "      <th>coef upper 95%</th>\n",
       "      <th>exp(coef) lower 95%</th>\n",
       "      <th>exp(coef) upper 95%</th>\n",
       "      <th>z</th>\n",
       "      <th>p</th>\n",
       "      <th>-log2(p)</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>param</th>\n",
       "      <th>covariate</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 rowspan=\"8\" valign=\"top\">lambda_</th>\n",
       "      <th>Intercept</th>\n",
       "      <td>46.049</td>\n",
       "      <td>9.973e+19</td>\n",
       "      <td>54.083</td>\n",
       "      <td>-59.951</td>\n",
       "      <td>152.049</td>\n",
       "      <td>0.000</td>\n",
       "      <td>1.081e+66</td>\n",
       "      <td>0.851</td>\n",
       "      <td>0.395</td>\n",
       "      <td>1.342</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>president[T.Bush 43]</th>\n",
       "      <td>0.369</td>\n",
       "      <td>1.446</td>\n",
       "      <td>0.399</td>\n",
       "      <td>-0.413</td>\n",
       "      <td>1.151</td>\n",
       "      <td>0.662</td>\n",
       "      <td>3.161</td>\n",
       "      <td>0.925</td>\n",
       "      <td>0.355</td>\n",
       "      <td>1.494</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>president[T.Carter]</th>\n",
       "      <td>-0.316</td>\n",
       "      <td>0.729</td>\n",
       "      <td>0.375</td>\n",
       "      <td>-1.051</td>\n",
       "      <td>0.419</td>\n",
       "      <td>0.350</td>\n",
       "      <td>1.521</td>\n",
       "      <td>-0.842</td>\n",
       "      <td>0.400</td>\n",
       "      <td>1.322</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>president[T.Clinton]</th>\n",
       "      <td>0.326</td>\n",
       "      <td>1.385</td>\n",
       "      <td>0.213</td>\n",
       "      <td>-0.092</td>\n",
       "      <td>0.744</td>\n",
       "      <td>0.912</td>\n",
       "      <td>2.104</td>\n",
       "      <td>1.527</td>\n",
       "      <td>0.127</td>\n",
       "      <td>2.980</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>president[T.Obama]</th>\n",
       "      <td>0.685</td>\n",
       "      <td>1.985</td>\n",
       "      <td>0.599</td>\n",
       "      <td>-0.488</td>\n",
       "      <td>1.859</td>\n",
       "      <td>0.614</td>\n",
       "      <td>6.418</td>\n",
       "      <td>1.145</td>\n",
       "      <td>0.252</td>\n",
       "      <td>1.986</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>president[T.Reagan]</th>\n",
       "      <td>0.077</td>\n",
       "      <td>1.080</td>\n",
       "      <td>0.251</td>\n",
       "      <td>-0.415</td>\n",
       "      <td>0.569</td>\n",
       "      <td>0.661</td>\n",
       "      <td>1.766</td>\n",
       "      <td>0.307</td>\n",
       "      <td>0.759</td>\n",
       "      <td>0.398</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>president[T.Trump]</th>\n",
       "      <td>0.605</td>\n",
       "      <td>1.831</td>\n",
       "      <td>0.784</td>\n",
       "      <td>-0.931</td>\n",
       "      <td>2.141</td>\n",
       "      <td>0.394</td>\n",
       "      <td>8.505</td>\n",
       "      <td>0.772</td>\n",
       "      <td>0.440</td>\n",
       "      <td>1.184</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>year</th>\n",
       "      <td>-0.019</td>\n",
       "      <td>0.981</td>\n",
       "      <td>0.027</td>\n",
       "      <td>-0.073</td>\n",
       "      <td>0.034</td>\n",
       "      <td>0.930</td>\n",
       "      <td>1.034</td>\n",
       "      <td>-0.715</td>\n",
       "      <td>0.474</td>\n",
       "      <td>1.076</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>rho_</th>\n",
       "      <th>Intercept</th>\n",
       "      <td>0.669</td>\n",
       "      <td>1.952</td>\n",
       "      <td>0.068</td>\n",
       "      <td>0.535</td>\n",
       "      <td>0.803</td>\n",
       "      <td>1.707</td>\n",
       "      <td>2.232</td>\n",
       "      <td>9.782</td>\n",
       "      <td>&lt;0.0005</td>\n",
       "      <td>72.648</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table><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",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Concordance</th>\n",
       "      <td>0.570</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>AIC</th>\n",
       "      <td>2680.513</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>log-likelihood ratio test</th>\n",
       "      <td>6.980 on 7 df</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>-log2(p) of ll-ratio test</th>\n",
       "      <td>1.214</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from lifelines import *\n",
    "\n",
    "wf = WeibullAFTFitter(penalizer=0.0)\n",
    "wf.fit(regression_df, 'T', 'event')\n",
    "wf.print_summary(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<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",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>model</th>\n",
       "      <td>lifelines.LogNormalAFTFitter</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>duration col</th>\n",
       "      <td>'T'</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>event col</th>\n",
       "      <td>'event'</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>number of observations</th>\n",
       "      <td>287</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>number of events observed</th>\n",
       "      <td>158</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>log-likelihood</th>\n",
       "      <td>-1338.519</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>time fit was run</th>\n",
       "      <td>2020-07-12 23:59:58 UTC</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div><table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th>coef</th>\n",
       "      <th>exp(coef)</th>\n",
       "      <th>se(coef)</th>\n",
       "      <th>coef lower 95%</th>\n",
       "      <th>coef upper 95%</th>\n",
       "      <th>exp(coef) lower 95%</th>\n",
       "      <th>exp(coef) upper 95%</th>\n",
       "      <th>z</th>\n",
       "      <th>p</th>\n",
       "      <th>-log2(p)</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>param</th>\n",
       "      <th>covariate</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 rowspan=\"8\" valign=\"top\">mu_</th>\n",
       "      <th>Intercept</th>\n",
       "      <td>57.104</td>\n",
       "      <td>6.308e+24</td>\n",
       "      <td>56.572</td>\n",
       "      <td>-53.774</td>\n",
       "      <td>167.982</td>\n",
       "      <td>0.000</td>\n",
       "      <td>8.988e+72</td>\n",
       "      <td>1.009</td>\n",
       "      <td>0.313</td>\n",
       "      <td>1.677</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>president[T.Bush 43]</th>\n",
       "      <td>0.503</td>\n",
       "      <td>1.654</td>\n",
       "      <td>0.432</td>\n",
       "      <td>-0.343</td>\n",
       "      <td>1.350</td>\n",
       "      <td>0.709</td>\n",
       "      <td>3.856</td>\n",
       "      <td>1.165</td>\n",
       "      <td>0.244</td>\n",
       "      <td>2.035</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>president[T.Carter]</th>\n",
       "      <td>-0.408</td>\n",
       "      <td>0.665</td>\n",
       "      <td>0.398</td>\n",
       "      <td>-1.189</td>\n",
       "      <td>0.372</td>\n",
       "      <td>0.305</td>\n",
       "      <td>1.451</td>\n",
       "      <td>-1.025</td>\n",
       "      <td>0.305</td>\n",
       "      <td>1.711</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>president[T.Clinton]</th>\n",
       "      <td>0.280</td>\n",
       "      <td>1.323</td>\n",
       "      <td>0.246</td>\n",
       "      <td>-0.202</td>\n",
       "      <td>0.762</td>\n",
       "      <td>0.817</td>\n",
       "      <td>2.144</td>\n",
       "      <td>1.139</td>\n",
       "      <td>0.255</td>\n",
       "      <td>1.973</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>president[T.Obama]</th>\n",
       "      <td>0.811</td>\n",
       "      <td>2.250</td>\n",
       "      <td>0.629</td>\n",
       "      <td>-0.423</td>\n",
       "      <td>2.044</td>\n",
       "      <td>0.655</td>\n",
       "      <td>7.725</td>\n",
       "      <td>1.288</td>\n",
       "      <td>0.198</td>\n",
       "      <td>2.339</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>president[T.Reagan]</th>\n",
       "      <td>0.017</td>\n",
       "      <td>1.017</td>\n",
       "      <td>0.278</td>\n",
       "      <td>-0.529</td>\n",
       "      <td>0.562</td>\n",
       "      <td>0.589</td>\n",
       "      <td>1.754</td>\n",
       "      <td>0.060</td>\n",
       "      <td>0.952</td>\n",
       "      <td>0.071</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>president[T.Trump]</th>\n",
       "      <td>0.684</td>\n",
       "      <td>1.982</td>\n",
       "      <td>0.815</td>\n",
       "      <td>-0.913</td>\n",
       "      <td>2.282</td>\n",
       "      <td>0.401</td>\n",
       "      <td>9.798</td>\n",
       "      <td>0.839</td>\n",
       "      <td>0.401</td>\n",
       "      <td>1.318</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>year</th>\n",
       "      <td>-0.025</td>\n",
       "      <td>0.975</td>\n",
       "      <td>0.028</td>\n",
       "      <td>-0.081</td>\n",
       "      <td>0.031</td>\n",
       "      <td>0.922</td>\n",
       "      <td>1.031</td>\n",
       "      <td>-0.883</td>\n",
       "      <td>0.377</td>\n",
       "      <td>1.407</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>sigma_</th>\n",
       "      <th>Intercept</th>\n",
       "      <td>-0.293</td>\n",
       "      <td>0.746</td>\n",
       "      <td>0.060</td>\n",
       "      <td>-0.411</td>\n",
       "      <td>-0.175</td>\n",
       "      <td>0.663</td>\n",
       "      <td>0.839</td>\n",
       "      <td>-4.869</td>\n",
       "      <td>&lt;0.0005</td>\n",
       "      <td>19.768</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table><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",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Concordance</th>\n",
       "      <td>0.587</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>AIC</th>\n",
       "      <td>2695.039</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>log-likelihood ratio test</th>\n",
       "      <td>6.228 on 7 df</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>-log2(p) of ll-ratio test</th>\n",
       "      <td>0.962</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "lnf = LogNormalAFTFitter(penalizer=0.0000)\n",
    "lnf.fit(regression_df, 'T', 'event')\n",
    "lnf.print_summary(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<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",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>model</th>\n",
       "      <td>lifelines.LogLogisticAFTFitter</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>duration col</th>\n",
       "      <td>'T'</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>event col</th>\n",
       "      <td>'event'</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>number of observations</th>\n",
       "      <td>287</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>number of events observed</th>\n",
       "      <td>158</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>log-likelihood</th>\n",
       "      <td>-1333.497</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>time fit was run</th>\n",
       "      <td>2020-07-12 23:59:58 UTC</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div><table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th>coef</th>\n",
       "      <th>exp(coef)</th>\n",
       "      <th>se(coef)</th>\n",
       "      <th>coef lower 95%</th>\n",
       "      <th>coef upper 95%</th>\n",
       "      <th>exp(coef) lower 95%</th>\n",
       "      <th>exp(coef) upper 95%</th>\n",
       "      <th>z</th>\n",
       "      <th>p</th>\n",
       "      <th>-log2(p)</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>param</th>\n",
       "      <th>covariate</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 rowspan=\"8\" valign=\"top\">alpha_</th>\n",
       "      <th>Intercept</th>\n",
       "      <td>7.282</td>\n",
       "      <td>1453.611</td>\n",
       "      <td>56.981</td>\n",
       "      <td>-104.399</td>\n",
       "      <td>118.963</td>\n",
       "      <td>0.000</td>\n",
       "      <td>4.622e+51</td>\n",
       "      <td>0.128</td>\n",
       "      <td>0.898</td>\n",
       "      <td>0.155</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>president[T.Bush 43]</th>\n",
       "      <td>0.103</td>\n",
       "      <td>1.108</td>\n",
       "      <td>0.425</td>\n",
       "      <td>-0.731</td>\n",
       "      <td>0.937</td>\n",
       "      <td>0.481</td>\n",
       "      <td>2.552</td>\n",
       "      <td>0.242</td>\n",
       "      <td>0.809</td>\n",
       "      <td>0.306</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>president[T.Carter]</th>\n",
       "      <td>-0.166</td>\n",
       "      <td>0.847</td>\n",
       "      <td>0.395</td>\n",
       "      <td>-0.940</td>\n",
       "      <td>0.608</td>\n",
       "      <td>0.390</td>\n",
       "      <td>1.837</td>\n",
       "      <td>-0.421</td>\n",
       "      <td>0.674</td>\n",
       "      <td>0.569</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>president[T.Clinton]</th>\n",
       "      <td>0.110</td>\n",
       "      <td>1.117</td>\n",
       "      <td>0.238</td>\n",
       "      <td>-0.356</td>\n",
       "      <td>0.577</td>\n",
       "      <td>0.700</td>\n",
       "      <td>1.781</td>\n",
       "      <td>0.464</td>\n",
       "      <td>0.643</td>\n",
       "      <td>0.638</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>president[T.Obama]</th>\n",
       "      <td>0.250</td>\n",
       "      <td>1.285</td>\n",
       "      <td>0.630</td>\n",
       "      <td>-0.985</td>\n",
       "      <td>1.486</td>\n",
       "      <td>0.373</td>\n",
       "      <td>4.420</td>\n",
       "      <td>0.397</td>\n",
       "      <td>0.691</td>\n",
       "      <td>0.533</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>president[T.Reagan]</th>\n",
       "      <td>0.149</td>\n",
       "      <td>1.161</td>\n",
       "      <td>0.265</td>\n",
       "      <td>-0.371</td>\n",
       "      <td>0.669</td>\n",
       "      <td>0.690</td>\n",
       "      <td>1.953</td>\n",
       "      <td>0.562</td>\n",
       "      <td>0.574</td>\n",
       "      <td>0.801</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>president[T.Trump]</th>\n",
       "      <td>-0.034</td>\n",
       "      <td>0.966</td>\n",
       "      <td>0.826</td>\n",
       "      <td>-1.653</td>\n",
       "      <td>1.585</td>\n",
       "      <td>0.191</td>\n",
       "      <td>4.878</td>\n",
       "      <td>-0.041</td>\n",
       "      <td>0.967</td>\n",
       "      <td>0.048</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>year</th>\n",
       "      <td>-0.000</td>\n",
       "      <td>1.000</td>\n",
       "      <td>0.029</td>\n",
       "      <td>-0.056</td>\n",
       "      <td>0.056</td>\n",
       "      <td>0.945</td>\n",
       "      <td>1.058</td>\n",
       "      <td>-0.002</td>\n",
       "      <td>0.999</td>\n",
       "      <td>0.002</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>beta_</th>\n",
       "      <th>Intercept</th>\n",
       "      <td>0.892</td>\n",
       "      <td>2.440</td>\n",
       "      <td>0.069</td>\n",
       "      <td>0.757</td>\n",
       "      <td>1.027</td>\n",
       "      <td>2.131</td>\n",
       "      <td>2.794</td>\n",
       "      <td>12.909</td>\n",
       "      <td>&lt;0.0005</td>\n",
       "      <td>124.239</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table><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",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Concordance</th>\n",
       "      <td>0.581</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>AIC</th>\n",
       "      <td>2684.993</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>log-likelihood ratio test</th>\n",
       "      <td>6.110 on 7 df</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>-log2(p) of ll-ratio test</th>\n",
       "      <td>0.924</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "llf = LogLogisticAFTFitter(penalizer=0.000)\n",
    "llf.fit(regression_df, 'T', 'event')\n",
    "llf.print_summary(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}