{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Living Arrangements Examples from CPS\n",
    "\n",
    "September 1, 2020\n",
    "\n",
    "Attempt at replication of [this](https://www.federalreserve.gov/econres/notes/feds-notes/an-early-evaluation-of-the-effects-of-the-pandemic-on-living-arrangements-and-household-formation-20200807.htm). \n",
    "\n",
    "---- \n",
    "\n",
    "**Update**: September 17, 2020\n",
    "\n",
    "The gap between the published results and my local calculations seems to be coming from how headship is defined. In the CPS, the household ID is the person ID of the wife in husband-wife households and the reference person in all other households. In the published data, the head of household is defined using a process that assigns an average headship to each person in the CPS.\n",
    "\n",
    "Currently trying to implement a new variable in the bd CPS called `HEAD` which is the average headship rate for an individual age 16 or older, based on Paciorek (2013, 2016). \n",
    "\n",
    "Also, because the published results are seasonally adjusted and the raw data have a clear seasonal pattern, it would be helpful to give an overview of the seasonal factors. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-09-18T01:11:41.038438Z",
     "start_time": "2020-09-18T01:11:40.850556Z"
    }
   },
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "comp_data = pd.read_csv('fed_hh_example.csv')\n",
    "\n",
    "import os\n",
    "os.chdir('/home/brian/Documents/CPS/data/')\n",
    "os.environ['X13PATH'] = '/home/brian/Documents/econ_data/micro/x13as/'\n",
    "\n",
    "import re, struct\n",
    "import numpy as np\n",
    "from statsmodels.tsa.x13 import x13_arima_analysis"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Aggregate approach"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 97,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-09-18T02:54:56.224716Z",
     "start_time": "2020-09-18T02:54:46.250243Z"
    }
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/brian/miniconda3/lib/python3.8/site-packages/statsmodels/tsa/x13.py:187: X13Warning: WARNING: At least one visually significant trading day peak has been\n",
      "          found in one or more of the estimated spectra.\n",
      "  warn(errors, X13Warning)\n"
     ]
    }
   ],
   "source": [
    "cols = ['QSTNUM', 'AGE', 'YEAR', 'MONTH', 'HHWGT', 'PWSSWGT']\n",
    "\n",
    "df = pd.concat([pd.read_feather(f'clean/cps{year}.ft', columns=cols)\n",
    "                  .query('AGE > 15') \n",
    "                for year in range(1996, 2021)])\n",
    "\n",
    "headship_rate = (lambda grp: grp.groupby('QSTNUM').HHWGT.first().sum()\n",
    "                 / grp.PWSSWGT.sum())\n",
    "\n",
    "data = (df.groupby(['YEAR', 'MONTH']).apply(headship_rate)).reset_index()\n",
    "data['DATE'] = pd.to_datetime(dict(year=data.YEAR, month=data.MONTH, day=1))\n",
    "data = data.set_index('DATE').drop(['YEAR', 'MONTH'], axis=1) * 100\n",
    "\n",
    "sm = x13_arima_analysis(data[0])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### \"Average\" approach"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 90,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-09-18T02:47:04.218571Z",
     "start_time": "2020-09-18T02:46:59.822246Z"
    }
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/brian/miniconda3/lib/python3.8/site-packages/statsmodels/tsa/x13.py:187: X13Warning: WARNING: At least one visually significant seasonal peak has been found\n",
      "          in the estimated spectrum of the regARIMA residuals.\n",
      "  warn(errors, X13Warning)\n"
     ]
    }
   ],
   "source": [
    "cols = ['QSTNUM', 'AGE', 'YEAR', 'MONTH', 'PWSSWGT', 'HEAD']\n",
    "\n",
    "df = pd.concat([pd.read_feather(f'clean/cps{year}.ft', columns=cols)\n",
    "                  .query('AGE > 15') \n",
    "                for year in range(1996, 2021)])\n",
    "\n",
    "headship_rate = (lambda grp: np.average(grp.HEAD, weights=grp.PWSSWGT))\n",
    "data = (df.groupby(['YEAR', 'MONTH']).apply(headship_rate)).reset_index()\n",
    "data['DATE'] = pd.to_datetime(dict(year=data.YEAR, month=data.MONTH, day=1))\n",
    "data = data.set_index('DATE').drop(['YEAR', 'MONTH'], axis=1) * 100\n",
    "\n",
    "sm = x13_arima_analysis(data[0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 98,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-09-18T02:55:03.469956Z",
     "start_time": "2020-09-18T02:55:03.216779Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "comp_data['DATE'] = pd.to_datetime(comp_data['date'])\n",
    "comb = (comp_data.set_index('DATE')\n",
    "                 .join(sm.seasadj)\n",
    "                 .rename({'headship': 'Paciorek', 'seasadj': 'Dew'}, axis=1))\n",
    "comb[['Paciorek', 'Dew']].plot(title='Aggregate headship rate, percent');"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 99,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-09-18T02:55:08.857435Z",
     "start_time": "2020-09-18T02:55:08.711597Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYIAAAEWCAYAAABrDZDcAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/d3fzzAAAACXBIWXMAAAsTAAALEwEAmpwYAABtdElEQVR4nO29eZjkVnnv/30l1V69d89Mz+KZ8YrH2xgPYGMgBmwwDolxgAC/BBwgcULgF8hyA2FJyE1y43ADTm4WiFmC4RIIgYANmBDbYMDG2IzBHs/YxjMejz1Le7qn91pVks79QzqqI5VUpeqq7q7lfJ6nn65SSSpJJZ33vDsxxiCRSCSS/kVZ7wOQSCQSyfoiBYFEIpH0OVIQSCQSSZ8jBYFEIpH0OVIQSCQSSZ8jBYFEIpH0OVIQSPoSIvoEEX1IeP8OIjpJRDkiGiOiy4nooPP+Net4qBLJqkMyj0DSixDREQAbARgATACPAvgcgJsZY5Zv3RiAJQCXMsYedpbdBeA2xtjfr+VxSyTrgdQIJL3MLzHGBgBsB3AjgPcC+HTAehsBJAEcEJZt972PDBFpK9lOIlkvpCCQ9DyMsUXG2G0A3gDgeiI6n4g+S0R/SURnA/i5s+oCEX2XiJ4EcDqAbzimoQQRDRHRp4loioiOO9uqAEBEv0FE9xLRTUQ0B+DDzjZ/S0TPOCanTxBRyln/CiI6RkR/SETTzj7fyo+XiFJE9FEiepqIFonoHmHbS4noR0S0QEQPE9EVa3YhJT2LFASSvoEx9gCAYwBeLCx7AsB5ztthxtjLGGNnAHgGtkaRZYyVAdwC28x0JoCLAbwCwG8Ku38BgMMANgD4KwB/A+BsALudbbYA+FNh/U0AhpzlbwfwT0Q04nz2twAuAfBCAKMA/hiARURbAHwLwF86y/8IwFeJaKKlCyPpe6QgkPQbJ2APopEhoo0AXgXgPYyxPGNsGsBNAN4o7pcx9g+MMQNACcBvAfh9xtgcY2wZwP/yrV8B8D8ZYxXG2O0AcgDOISIFwNsAvJsxdpwxZjLGfuQIo18HcDtj7HbGmMUYuwPAXgDXrOA6SCQu0pYp6Te2AJhrcpvtAGIApoiIL1MAHBXWEV9PAEgDeFBYnwCowjqzjtDgFABkAYzD9lc8GXIcryeiXxKWxQB8r5mTkUj8SEEg6RuI6HmwBcE9sE05UTkKoAxg3Dd4i4jhd6cAFAGcxxg73uRhnoKtUZwB4OGA4/g8Y+y3mtynRFIXaRqS9DxENEhErwbwJQD/lzH2SDPbM8amAPw3gI86+1KI6Awi+oWQ9S0AnwRwExFtcI5hCxG9MsJ3WQA+A+BjRLSZiFQiuoyIEgD+L4BfIqJXOsuTjuN5azPnI5H4kYJA0st8g4iWYc+kPwDgYwDeWn+TUN4CIA47H2EewFcATNZZ/70ADgH4MREtAbgTwDkRv+uPADwC4CewzVh/A0BhjB0FcC2A9wOYgX1e/wPyOZa0iEwok0gkkj5HziQkEomkz5GCQCKRSPocKQgkEomkz5GCQCKRSPqcrswjGB8fZzt27Fjvw5BIJJKu4sEHHzzFGKspSdKVgmDHjh3Yu3fveh+GRCKRdBVE9HTQcmkakkgkkj5HCgKJRCLpc6QgkEgkkj5HCgKJRCLpc6QgkEgkkj5HCgKJRCLpc6QgkEgkkj5HCoIu58Gn5/HTZ+bX+zAkEkkX05UJZRIb3bDw2o//CABw5MZfXOejkUgk3YrUCLqYb++fWu9DkEgkPYAUBF3MFx94BgAwnI6t85FIJJJuRgqCLubEQgkAsFSswDAtlCrmOh+RRCLpRqQg6GJ0wwIAWAz42B1P4OUf/T4sS7YelUh6ia/97Bh+5/MPrup3SEHQxZQNE+m4CgC47/Asji8UcWgmt85HJZFI2smDT8/jBwdnVvU7pCDoYnTDwsbBJADg4ElbADz4tAwllUh6iXLFQsW0VvU7pCDoYnTTwsRAAgCQKxsApCCQSHqNsmGhYrIVmX0ZY7jz0ZMwGggSKQi6FMtiqJjM1Qg4MrlMIuktuC+wYjWvFTz49Dx+83N78b9uf7zuelIQdCm6I+E3OhoBAAylYjg8k8dyqbJehyWRSNpM2bCjASvmCjQC5//Xfnas7npSEHQpZWeWMCEIgou2DQMATi6V1+OQJBLJKsCf9YoRXSM4OlfAn966H+WKvc18oVLXtCQFQZfCZwnphIaBhF0p5IItgwCA6eXSuh2XRBIF02L4/H1HZO5LBFzTUBMO4x8cnMHn7nsaT8/l3WUHTiyFrt8WQUBEVxPRz4noEBG9L+BzIqL/43y+j4ieG3VbSTD85kioCoaczOILtgwDAKalRiBZJfYfX8SJhWLT2y0UdBw5VR2UvvLgUXzo1gP49D1PtfPwehKuEZSb0AgMx4yUKxnusn3HF0LXb1kQEJEK4J8AvArALgBvIqJdvtVeBeAs5+8GAB9vYltJAFwQxDUFI+k4AODCrUMAgJNLUiOQrA6v/od78MIbv9v0dn9350H82qfud98fn7eFSTODW79S9RE0IQgcMxCPJgSAUiV8+3ZoBM8HcIgxdpgxpgP4EoBrfetcC+BzzObHAIaJaDLitpIA+AOU0BQMp2NIaAomh5JIx1VML0uNQNJZzOTKeHap5NqpF4p2QMNQStbJaoTrI2jCWczDRZcFjUCvI3TbIQi2ADgqvD/mLIuyTpRtAQBEdAMR7SWivTMzq5tl1w2IGsHkUBI7xjIgImwcTEqNQNJx5MsGTIth0REAS87/VExdz8NqyGyujHsPnVrXY1iJjyBII1htQUABy/yiK2ydKNvaCxm7mTG2hzG2Z2JioslD7D14+GhcU/D+a87Fv771eQCADQMJqRFIOg5uq57N2/cmFwirnTHbKl984Blc/5kH1rWGV7t8BNzEFEQ7BMExANuE91sBnIi4TpRtJQHwsLCEpmI4Hcfm4RQAYONgEtNSI5B0GHxmOpvTAVQFQb3BqRNYLhswLLaiZK52Ua6sxEdgr7uWGsFPAJxFRDuJKA7gjQBu861zG4C3ONFDlwJYZIxNRdxWEoBu2jdHXPP+hBsGEji5VAZjsgqpZG0wLYbjDSKJ+IA0l/cJgjoOzE6AH5+xgmSuth3DCkxD3J/Ar3sqptbVKFoWBIwxA8C7AHwHwGMAvswYO0BEv0NEv+OsdjuAwwAOAfgkgN+tt22rx9QPuD4C1fsTbhxMolgxPTMBiWQ1+fQ9h3H5jd/FoenwyreuRuAIgvkC1wg6XBA4Gst6CQLTYq69vxlBYPo0gmxSq6sRtKVnMWPsdtiDvbjsE8JrBuCdUbeVNMaNGop5BQHPNJ5ZLmMgKSMyJKvPw8cWAQD7ji3gzA3Zms8ZY8gLGoFuWK5m0OmmIR5yuV6mIXHwrjeQ+6kIPgJVIUcjWF0fgWQdKIdoBIMpW7aLYWMSyWqyecgufDi1GOyb4tUzATsK51Su7Pmsk1lvjUAcvPUmjsEUoobiqoKEprgBJkFIQdCl6CEaQTZhawFBguCZ2QKOzhVW/+AkfQXPBQjzE+QFM+VsXvdEtXW6j8DVCNYpuknUApqpNSQ6i2MqIa4pda+1FARdimsaUr1x2ANJWyPIlWsrkL73q/vwga/vX/2Dk/Q84sDIB6uw0hOiv2our2NGEAT1ZqmdgKsRrFP4qKgxrcRZDABxTW2oEbTFRyBZe8SEMpGsU4BuKUAjWChWOj5uW9K5iLH0pYqJmGOW5IOVWEtIhAsCVSHM5XUsFGz/QEJTusZHYK6Tj8BrGmrGWVz9rRKaIjWCXiVMEAw6DuJcgCAoVUycypVxaHoZ39o31bZjOTS9jGPz0uTU65hCSHJRqBrKBcHR+WLgRIPfi5uHk5jN627o6MbBZMebhlrpBdAOxPpAzTmLq+vapiEVZekj6D3KhglNIaiKNzk7k7BNRUE+gqJuYqFQwT9/70n8/pcfalu25Lv+7Wf4y28+1pZ9SToXcZYpDuB8sDQthmcCfFB53b4Xt49mMOcIAiJgLBvveGdxaZ3zCEQtoLlaQ6JpyHEWr3JCmWQd0A2rRhsAAE1VkI6rgT6CkvPAPnxsAbpheaI3VoplMRw+lcdMG/Yl6WysMI1AEApTC7WRQ3xSsm00DdNiODpXwGAyhqRWP6SxE3A1gvUyDVVW5iMQfRpxbhqS4aO9R9mwkAgQBIDtJwjTCADgsGPLPTrffF15P9PLZeiG5ar7kt5F1Aj4vQTY9yLXTGdytYIgX7bX3T6WBgA8dSqPoVQMiZgiNYIGeHwEK4gaAlANH5UaQe8RphEAduTQsi+z2LKY+9DxiV2jsgBR4KaAhYIUBL2OOCkueXwEJraO2LWuZgIKHnLtdPuoLQgOzziCoIEDsxPgdX6MTggfXaGzOOYIglUtMSFZH3QzXBBkk7EajaAUoBa2w8HLBcFSsSLrG/U49ZzFo5k4kjElRBDY625zBMFy2XAEgdrx4aMlXuenA8JHm7lWotCwfQSq1Ah6kbJhIqEF13IfSGjIlbwzdFGV5xxrg2mICwLdtOp2QJJ0P6YvfJRTrlhIaiomBhLBgqBkIJvQMJaNu8uqGkHn+ggYY+7guV4awUrzCERTVkL6CHoX3bBqyktwBpK1PoKSbzYQ15S2CAIxU3mhqLe8P0nnIjqLRaFfMkwkYgomsonAoIFcuYJMQsVopioIBrvAR+AdhNfXR6BQsz6CWtOQ1Ah6kHIdH0E2odVUH/VrBLu3DeN4G01DAKTDuMfxOIt9GkFCUzAxkMD0Uq0gyJdNZBMaEpqKASfhcSgVQ1ytXxp5vRH9F8Y6F53LJLTmwkctr2koriqoZ92SgqBLqRc1NBDkIxAe3ISm4MItQzg2X2zZrn90roAtTlMc6TDubcKjhmwz5YaBZKBGMJfX3XpEo455qBo11LmmIdGvtn5RQ/aAPpDQmvIRePIIVKWmJpkfKQi6lHpRQ9mkrRFYATO4kXQMO8YymBhIoGxYyAf4DqLCGMNsXsdZG+3Sw1Ij6G08PgLD6yzmGsFCoVIzuB9fKGLLiO0o5uYh7iOomMyz305ipTH8q3EMmYTWZNE5Xx5BiBmZIwVBl6LX0QgGeeE5vaoV8BncX113Af7lzZcgHbcdzQV95eWql0p2Q/Kd4xkAwKLUCHoaMWqo5MsjSMQUtxcGb0cJ2GHLU4tFbB62S1WPeQSBfQ82Y/teSzwawbpFDdkVBJIxtUlnsVhiQkE8JLCEIwVBl1IvaogXnhPNQ9w0dNpoGjvGM0jF7XWCoomiMu80F3EFgdQIepogDROwY+0TmoqJbLUpEmcmV0bFZNjqmA/9GgGwckFw/+FZ/PDgzIq2jYLXR7BOJSacCV+8QfVQP4av6FzYpJHTkiAgolEiuoOIDjr/R0LWu5qIfk5Eh4jofcLyDxPRcSJ6yPm7ppXj6Sfq5REMBBSe4w9uytEEqhrBygXBnFNFcttIGqpCMmqoxzFDooZE0xDgFQQ8aXGLk3A25ggL7iOwt1/ZPfibt+zFmz/9AN77lX34wNceWdE+6uH1Eaxf+GhcUxBTCRVj5bWGwsYKTqsawfsA3MUYOwvAXc57D0SkAvgnAK8CsAvAm4hol7DKTYyx3c6fbFkZkXrho9kk1wiqM3SuESRjtgBItUEQcI1gNBPHUComNYIeJyhqyDAtGBZDMqYKvTCqE5DjTojyZkcjCDINrTRyaDhjT3j+fe9RfOH+Z9qe0OjRCNYxfDShqYipzWoEXtPQqmoEAK4FcIvz+hYArwlY5/kADjHGDjPGdABfcraTtAC3ywaRCRjkuQko5QiCtPO/FdPQnCAIhlMxGTXU44gRlFwQ8MEpoSmuSdIjCLhG4AiCq3ZtxFsv34EtIyl3lrpSjUBTFFx+5hhef8lWAGh7QqMYabeePYt59dBWi87Vo1VBsJExNgUAzv8NAetsAXBUeH/MWcZ5FxHtI6LPhJmWAICIbiCivUS0d2Zm9eyC3ULFsNzGIH6CZvtF5yFxBYHjI2jFWTzvmIZGMnEMSo2g5wlyFvNZc0JTkHEEgdia8sRCEQNJzTVXbh/L4M9+6TyoCrmz1JUO4PmygW0jaVy4bRgAsBxQcbcVRE1lvTSCXNlAOu5oBM1EDfnDR1t1FhPRnUS0P+Av6qyeApbxo/w4gDMA7AYwBeCjYTthjN3MGNvDGNszMTER8at7l4rFoKlBl7Y62IszGj6D4w8fFxbFFlL85/IVxFUFmbiKoVQMS1IQ9DTcNKQQsP/EIoq6WW2ZGlORjqsg8gqC4/NFVxvwk3A1gpUJgqJuIhWvJqkFNWNqBfH5WS8fwcxyGRMDCcTU5jQCf62hljUCxtiVjLHzA/5uBXCSiCYBwPk/HbCLYwC2Ce+3Ajjh7PskY8xkjFkAPgnbjCSJgGFaiCnBP191tu+N7EjGFChOueB2OIvn8zpGMjEQEdJxtSWhIul8eImJd770TJxcKuMz9z7lmnUSmgIiQiauuUXmALtM+cbBZOD+qj6C5u8bxhjyuoFMXAs0SbUDT4mJdYoamlkuY8NA0hEE0Y9B9OfE18BHcBuA653X1wO4NWCdnwA4i4h2ElEcwBud7bjw4FwHQHZWj4BlMVgMDTWCok8j4I5ioH1RQyPpuPudrexL0vnwweWy08fw4rPG8ZUHj1U1AmdQzyRUj0ZQNkz3fvTDfVwrCR8tGxYsBqQTqhsc0WsaAWMMMzlbI2gmfJQxBsNi7nWPr3b4KIAbAVxFRAcBXOW8BxFtJqLbnYMyALwLwHcAPAbgy4yxA872HyGiR4hoH4CXAvj9Fo+nL+COIE0JEQTc7OPY///9J89g37FFzwPpX6ceT8/m8aUHnqlZPp/X3bjwVFz1PDiS3oPnESgK4fwtQzg6V3CFPx9oMnHNk8hYL6iBb/PZHx3xFC+MAhc2okaw1GZBwIVcXGtuNt4uFosVVExmCwKVIgtMLrB521pehroeWisHyhibBfDygOUnAFwjvL8dQE1oKGPsza18f7/CQ8O0EGdxTLV7GRcrJubzOt77VTvG+nQn8Quw1UVVoUiz+K88eAz/8N1DeN0lWz3fOVfQce7kIABbI2glAknS+YgTkNNG0zAshqdn7W53fLDPJDQUBI2gXpgzH5zu/vkMbrrzCXzsV3dHPhZ+36bjKgZ53kybTUN8YpOJq+tSdI7nY0wMJDC1UIzsI+C/k20i1hFXVz9qSLIO8NlJmEZAREg7ppqfHJlzl4umIXGdIA6cWMSh6WUA1YfO79Sbz+sYdUxD6biKQsWUzWl6GB41pDiCAACeOGnfI17TUG35iSBEc0WY+SiMqiDQBNNQ+6OG7GQuZdWihsqGiTsePRn4GRcEGwYSiDURPsrX4+bf2BqEj0rWAW6vDAsfBYCkY6p54KmqIIj5fAqpePgs/v1f248Pft122QQJAstiWChWMJKOud/H2MojQCSdDzcNqVQVBAdP5gBUB3V/CXRbIwge5CeHknjT8+04knyTs/m8Y35KJ1TXBLIaGkHSEQSrZRr67wMn8Vuf24unnD7iItOCRhB3jiHKRMt0NQLHNLQGzmLJOuCq6CHOYsCZoesmHjgyB644+HsU81l8ELO5svuQcxVZjO7I6wYYq5azSLUhQU3S2fABRlUIk0NJqArh0LQjCATTUF73OovDNAJNVfDXv3Ihzt8y2HQOSqHMzTZ2n4O4ptT06fZzeCaHh48u4BsPn8BL//buhjZ3+9hVaCqtmmlowcnF4Tk5IqJpiM/ooziMudDi0YNrkVAmWQe46hcWPgrYA/N8oYIDJ5bwil2bAACnct6bLRXXQp3Fi4UKZvM65vK6O7iLKfdc/edJROk25CVIOhsePqoQQVMVbBlO4bAzk62ahjR3dm9ZDBWTNSyBPJhsPhnR1Qic+85uz1pfELzso9/Htf90Lx54ag5PncrjyZlc3fWLuol0XIWq0KoVnePCK+jYZ3JlJDQFAwnNNetGmWhxoXXRtiHs2T6CMyYy0BTCeZsHQ7eRgqAL4fbKehpBKq7i2HwBpsXwkrODE/C41uDHtJh7gx6azrlag1iEi6vhXC3nN6oMIe1d+GRUdVRMbh4Cgk1DfPbaaDY6lIo1HfHDB0Q+EeE9OKJw+JQtAB6bWqr/HRU79DWmKKsWPsoFQNCx82QyIqomzdU5R8Zs0xEfH7aPZfCVd7wQw+k4iAjf+r0Xh27bUtSQZH3gEl8NcRYD9iD/+JTtyBvNxPHHV5+DMyeyNev4O5n9/Z0Hsfu0Yff9wenlmnICQNWmm3U1Avu/DCHtXbizmE/wt4+lcc8h+7UYPlqqWDBMS8gxaCwIWtUIshE0As6DT88DAB5/drnuesWKhVRchW5Yq+YsztXTCBxBAFQFXpggKOgG3vzpBzCaieP915wLoNYnWA+pEXQhXE2t5yxOxaqZvsPpGH73ijPxivM21a4jzOBNi+GmO5/ALT864i6zNQL75hMdwVWNQHP3BUjTUDdxdK7QVDKXm0dA9gDzlst2uJ9xjZBriHnddPe9GoKA+whEQdDIR8BNVLy2UUONQDeQiqnQVGXVMou5AAg69uWy4frgeGRUmFP9/f/5CB58eh53PHrS1V7UOqZjP1IQdCFGg/BRAG7jGcAWBEHYzuLqjcUL0B2crs6UDk3nqj6CANMQ1whSccXZhxQE3UCpYuIVN/0A//Hg0cYrO4jOYgA4Z9MA3vi8bYhriiAIqsUMq+Un6oeGDqZi0A2rRpu849GT+Jv/etzTEIdT1Qjs7xtI2hrBUqmCv/zmo4HFFCeHq6UuYirhsalGGgE3DdGqmYbq+QjKTtQSAGQdAevX4Dn7ji+6r3n72Vid8cGPFARdSCVC+GhKiNQYTsWD14lrHo2Avz7m1JDfMJDAiYViiLPYJwhirXc8k6wdyyUDxYrpaSLTCJN5NQIAuPG1F+LAn7/SFQ5iBVLdiOYjGHQa24tFC7/64DH81uf24uN3P+mGUYoUdLt2Fv9e7pu48duP41P3PIX/2v9szTZDqeqE6NLTx3AqV8ZsLvz8i7qJZNyJGmrCNDSX1/Hn3zgQSduq+ghqNSLdsJCIcY3HPvZTOR3/cNdB6IaFp07l3bFgNqe7zyLP0g5LOA1CCoIuJFr4aGONIONzFvOZBA9VPm007Q4YgNc0lPebhtyoofbGcktWBz5jbsaUZ/k0Ao44Icm6Mf1mUz4CwNvqdK9jx/cvF48/I9zj3Fl835OzNcfEEXMBLto6DCB8hg04UUMxuwR0M/0I7jl0Cv967xE32a4ero8gwORTqphV34tzXb9z4Fl89I4n8N3HT+KVN/0AX//ZcVRMC4vFCs7fYkcFPcMFgdQIehs+C9DqhY/yZBJBbffDK4byJBW/Or11JIXlkuEKC1F15xUms/7wUV0mlHUDPPy31IQG16jGFQB3cG5KI3Ds3+KAP7Nccl8vBWQMF8qme48D9ox5Lq+7iVlB24iZudvH0p5zCqJYsb9DU5rTCHiJjXr7/qfvHcI5H/y2IAhqf4eyYSHpaPYDjkbAtfXHppahmxZmcmW3U+AFW4YAiBqBFAQ9Db8p60UFcOftcCpYGwBs0xBjVeeZ376/bTSNYqU6s/NrBArBvVGTK3AWf/6+I3jlTT+IvL6kfXCh30xTGEsoMRGGGN3ir0waBtcIxMF7ernsZq0vBnS+y/s0At4mkxOkRXBBcNWuje4Epl6iGPcRaE32AuDPUb1t/vd3fo6yYbmmuaDyGLZG4HXCH3MGee7HK+qmmx90viMIqhqBNA31NI2KzgHVGXqYWUhchw8KYkRCKqZi3Gk0zvE7izMJDeTYi6uZxdFNQ4emc25Mt2Rt4WbAZgS3KZSYCCO7Ah9BkGloZrmMMzfY4c6BGoFuIp2oCpirz9+Ea3dvxuuctpVLxdr70DAZXnfJVnzyLXvcZydspm9ZDCUnfDSmNpdQxq9ppY6PgE+g6pmGRI1AUxUkY9XsaZ71X9BNzOZtYTI5lMJ4Nl4VBFIj6G0aFZ0DqjP0MEcxYLeYBIB7Hbuq6OgdSsUwmPLOsvwaAX/oAfth15yKp1EpVazI9VMkjbEshlf/ww/x7UemGq7LzRcrEQTRNQJvieowXEHgzPwti3kEQdDsPl/2agRnTGTx92+8GH/7+oswnk0EbqOb1faufJAMm7Xz5MlUTIXaZEIZn1DVKwfhfy79vgrDtGBYzKNNic/bEafqa0E33d7hY9k4JodSmFq0zWrSR9AD/PPdh/Ab//pA4GdmxFpDADBURyO4+rxNuPi0Ybzvq/swvVRyZ4mArUlwuyRHjBriGoFIs81p+MMWteGGpD453cD+40ueUMIw8gF+n0ZYLNhZLMIH9YVCpfmoIWcwnC/oMCyGM5wEyKDZfUH3+gi8+9NCfQRx55nhg2TYTJ9PilJxO3y0maJzVdNQ+DZ+Td2vEQQ52kVBwPdd1A3XNDSWiWPzcLI6PkjTUPfz82eXsT/kgY7kLI7gI4hrCt539XNQ0E089uyyx6wzmIq5DyjHX2LCLwh4xdOo8HXXo+lHL8JDEaP0jnajhpoQ3G6JiTqmobhTG2e+oEcuMRFTFaTjqjuLn3FCOjcPp5ARlouUKuGdz8L6Z1cMyzUJ8WcnTCPgmlJyBUXnihF8BKIgSMfVUEEgBnr4nzfAMQ3lytAUwmAyhsmhan/oNTMNEdEoEd1BRAed/yMh632GiKaJaP9Ktu9HKqYVOruO5CzmGkEdQQAAm4bsJJuZ5bJHIwg0DfnyCAZ8N2a6TlnrIIrO/urZUiXR4SaJKHV7eNRQU+GjrrO4/nrDmRjm87p7vzQyDQG2s3fZmcVPL1Wrbg6mYoGz+7Jhhe53MBkiCEzmmob4s2M20AjScdtZHLZeEDzZrWJamM2VA02fYnjr5FASuZLhWY9PksRzDBIExYptGhrJxKEohC3DVUGwliUm3gfgLsbYWQDuct4H8VkAV7ewfd+hG5YntFMkmrPYvmnqOYsBuA7hU7myK3i2j6WxeSjpdn7ieMpQl003koHTtGmo0njmJIkOn1U2pRG02VkMACPpOOYLFZQjagQAr1pqH8u00JAlrPyE7jSNCcK/zbcfmcLDRxdQsQTTUB1n8bv+7af4o/94GADczOJmtFYuRE4ulXDZX38X3/v5dM06or9tcigFw2KeZUEagX/ixb/rVE7HmOPvE7Onmykx0WrRuWsBXOG8vgXA3QDe61+JMfYDItqx0u37Ed1kbqMXfx4AvynrpZBztXkoHe4sBuwHMBVTccp5+FIxFV/8rUuRTWrw791fa6jGRxBXmxpYygGJapKVwwfS5QidulaSR+AvMRHGcDqOhYLu/r6NwkcBO/+Az6TFOvxhJarrCQLbR1DVit7xhZ+6r11nsRLuLN5/fBFHZu3IGx4+2oyzuOAKgjJ008Lx+WLNOuI9z7Xy5ZLhPutBjvYw05DFyu6EbrOgEayls3gjY2wKAJz/G9Z4+56Fm0uCTC38pqynEWwcTCCmEnaOZULX4YwPxDGTK6NQsWf5m4dTGEzGkIlrEO8lT9SQ7o0aAprvW8xj2KVG0B5yTZiG3DyCZorOMQYiuCHDYYykY5gvVFwfQRTTkN3i0j6m6eUSsgkN6bhmm4YCBIFtGgr3ESwWK2CM1fisYpo3aijIWTybr/bt4CUmmik65w/HFk2uzy6WcGKh6ApJwDYNAV6HMX82xKY+vPDcpsHqrL9YMTGb1zHqaASbh0TTUBudxUR0JxHtD/i7NvK3tAEiuoGI9hLR3pmZmbX86nWBP0RBHcSMCDOzDYNJ7P3gVbj8zLGG3zWRTdimobLhicRQFKpWP0xonps3HxI11IxGUJTO4oYwxnDtP92Lbzx8ouG6+aZMQ46PoEmNoJFZCOCmoaqPoFFjGoDXCrKPZS6vYyxrD2zc8fvzZ5fx6n/4IR4+ugDGGHSzjkaQjMG0GAq6iWcXS57PYg2cxRXT8oRypuPN9yPg15bH/BfKBn54cAZH5wr44NcfwXu/us9Th4hrBGLhOf6sJQPCR8/YUJ3cFXTD7h3uCIKJgYSrCTTS3EQa/kKMsSsZY+cH/N0K4CQRTQKA87/WGFafyNszxm5mjO1hjO2ZmAhutNJL8Bs0UCOwGjuLAfshajR7A2w/wallHQXd9MRmA3AdxsPpmDt7LBsmKiar1QiaNA3x2VozpZD7DdNiePjoAh6JEBJa1Qia8xFEzeMwLRZpcBlOx5zSJAZiKtXNO+CInc2WS4abKTyY0nBisYRf+ed7sf/4En54cKahpiEmqJ3wtWflz0yYs9jfMjLlRA1ZDIFVUIPggoCfT0E38c4v/BSfufcpzBcqmM3pKDumreF0DNtG0s55V383/qx5NAIuCJywWlUh5MsmlkqGe86qQtjoaAxr6Sy+DcD1zuvrAdy6xtv3LHxwDArHNCKEjzbD+EDCNg0FxGZzh/FwOubOUvjMJeNbNxOP3hwEEASBNA2FwrWlKN23+DqlitVQuOaF2jZRfTRRBQGfnU4vlyP5BwBbEHDhtFyquDksfIDj5pWpxVLDYnaDQskKf5/uakJZsLN4Pu8Voimn6ByAyIXn/KahXNnAUslA0enRwEt0v+6SrXjg/Ve69n3RFxLkX+Ea+FlOot3m4aT7m4tBITxyaC2rj94I4CoiOgjgKuc9iGgzEd3OVyKiLwK4D8A5RHSMiN5eb3uJYBoK0AgqEcJHm2E8m8B8QcdyqVKjEfCZ2XAq7j6APKpjzFeCYsOgbWKKGmrHZz3SRxAOH3zCGpKIiOs0chiLBQajmodMFs00NOwEKJxcKkWKGALsSUVO0Ai4JsonIjvHM7ho6xCeEZrp1IsaAuxM5RMLtmmICw03fJQ7i32DOy/XwEk5PYuB8HIUflzTUMnr/C4bliMITJQrFpKairimuEmfHkHgRg2J5eTt9V581gT+9vUX4dqLtlQ/EwQBjxxqxlncUtQQY2wWwMsDlp8AcI3w/k3NbC8RTEOBPgILqkKRzD5RmBhIgDHg6HwRl2xPej4bTMaQ0BSk4qrrRDviVHjcOe51RG8cTMJidijqxkHvfvyYFnMfaGkaCocHDUTRtERBsFQyagS1Z12xD0XFRJQEHstikcw8vFjcyaVyJP8AYM92eYvLpWLF9U3xc7ry3A2YWizhkeOLVUEQsm8uPJZKhmsa4hOrWIPwUb9GkIyp1SzkCILAFMJAeRTUSaeSqm5YKBsmirpdyJGbfYJqLZUCNIJXXbAJyZiKHeMZ7BjP4FM/POx+Jpas2DKcgkKyxERPUDGqKeR+DJM19SM3YsJxzM3l9Rpzz2AqhlRcRUJT3JA2Hlq3I0AQAKhx0AUh5iRIjSAc7g+KZhqqXtNGDuNC2XBn1FH9OiaLZhoacTSCZxdLHht3Pdxidbrp8RFc99wtePWFk/j/X34WThtN4/h80Z1xh+2bO1+fns3jxKItCLgbhAsPNSR8dE7wEcRUQkxVIpmGvvf4NP757kMeTcvNi1jyagR53YBuWkKfZ1vrCNIIxHNMxzX84oWTnvccsZTMb7xwB/75154rG9P0AnodjUDMkGwHYpXRlM809Kbnb8MfvuIcJDTVjQI5ciqP8WyixlnMw9qeXWosCMTyx1IQhKP7ZpdhnFgoerpcNXIY53UT444tP2pZENPydicLg5spihWzKY0AsAVeTq/26t06ksY//n/PxWAyhtNG0zAshqedgmtxNdj/MDGQwORQEvuOLYb6CGIh4aO8tr+qULVwo3M+tz4UHrn1nz87js/c85THzMa1uFM5LghM6KYF/pV8tk9ENUlwQRqBn7QwaRNLyWwYTOLq8yeDNglFCoIOhZsEgnwE3DTULs7ckK22GvRpBJdsH8WbL92ORKyqETw1m8fO8XTNfjYO2QLlZCRBUD0vvQfDR+967CQ+f9+RlvfDB6p8QOMSzoNPz+OFN34X3zlw0hXOQYXaRAp61XQUVRBYFkOUcX1ESGKMqhHwQW16qQTGqs1qRE4bte+5Q9N2CeZ6/oeLtg7jJ0fm3EYuHM0tOmdv6/dnzeV1DCQ1jGXi7jFdc8EkXrFrI/7im4/iGUcb9rNUrHiaOAHVyRz/Cts0VJ30iM7uoVTMk/8RpbubmGg63CBxtBFSEHQoep3wUVsjaJ8gGE7H8bwdtpU4HVLRMaEpHo1ge0Ci2njGjmGOYhryCIIe9BF89afH8Jl7j7S8H64t1TMNiYJ30s1SDdcIdMMu/81j9aN2lYvqLE7HVde0E1Uj4AKM3zv+RjOA3SgJqAqCeoPkhduGMLVYgm5YeMHOUXd53KcR1JiGnJj8iYGEm50fUxX8+qXbAQAzueB7e6lUQdmwArOgOWW/IBCE5KBPI4giCMRnNUhwNoMUBB0IT5gBggWBaVltCx3lvORsOzcjzKyT0FSUDQv5soHp5XKNoxiwE9A2DCRwcqlxQ3TR5BVmGnrTzT/Gh287EOXwOw7DZE0VKguDX5t6UUPigDDphA7WMw1xO/ZYxtYIovoIojqLiQjnbrL75zYTPgoAJ1xBUFsji/ug+Cy/kUYA2NruZWdUkyp5ZjERQQ1oQTlfsAXB5FDScwz8u8JCbd3Kqcvh9z73EXDEa+M3DZWdfsX1AkL47z6Y1JryBwQhBUEHYlrMdW4FRg2ZrKkSs1G4xrEpnr1xIPDzZEyBblr4udOQe0dI6YoNg8mIpqHGPoL7Ds/isz860nBfncBiseIZ+A2rXYKA95M2QxOaxMFlw0ACqkJ1TUM8YmicawRtdhYDwLmT9n0U9T6tagT2IB+kEcQ1BXFVwYIzYNYTBBduHcJAQsP1L9zhMaGIvjVVoRoH8Fxex2g6jg/84i78zWsvdJcnGggCfr2n6wgCf0/wGtOQTyMI6zXO4Tk/rZqFACkIOhIxwSrQWWy111kM2BFAez94Jd56+c7Az/ns5Ws/PY6YSnihMMsS2TSYjOQsLjcwDYnZrs2k968HZcPEi/7mu/j6z467ywyLuWWbW0E89zCHsTg4ZRMaBpLBjVk4C05kzMSA4yOImkcQscQEAOzabGsE0xG0Q6A6u33WWT9IIwDswY8ffz2zyUAyhgc+cCV+/QWnISmsJ5pUYz6NgDGGk0sljGbi2Dmecc/B/i6nGFxAj2fGmDuITy+HdwfzdyHzCgKtxlncqEZT1ArDUZCCoAPhoaNAeNG5doaPcsazidAZH78p/+PBo3jZcza4bS79bBpK4mQUH4FR61QTEQe3J2fyDfe3nhSdkEdRAJqW1VSf2zDEaxPmMBb9LXFNQSau1S0HPut0tNrqlDaIrBFEzCwGgHMn7UH0yZloPamjaASAHcwwH0EQALbQICIkhJm16LPwVxU9Nl/EqZyOC7YO1eyLax9h9ypfzjWCoD4gfr+NeFxioTy+z0aO9nTEniNRkIKgA2moEZisZZtgs4xk7JutVLFw3cVbw9dLx7FcNhrO4kUHpSj4OOLs6NGpxnV21hM+4IvCq2KyyLVp6u5bmLGGOYzF753N6UjF1RozhAjPnt02avsTooePRhcE3MQYVRhyH8FUHWcxYA/uJbeYXTT/gygwxOfG35T+wafnAQCXbK9Nr3NNQwHXSjTpTDuTgaDB2V9c0W8aMi3mmu3KhukpOBdEO01DrfYjkKwCoiAICx9tZ9RQFH75oi0YyyQwmy/jFbs2hq7Hm9XkdRNDqXBh5Q0frf9wHTi+hOsuXslRrw18sBZNXKbFYLbBNFTxaAQRBEG+jEy8foOgU8uORjDcnEZgMRYpjwCwQxtff8lWvOis8Ujrp+MqiKpRQ/6mSByx4m3U0FSvj6B6/JqieATt3qfnkE1oeM6mQfjh3xXkIxAnLdyRPT6QwOFT9TVZvyDg+8o6WdaNzi9KO9qoSI2gAxFbNwYXnYs+M2sXqkJ4ydkTuO7irXUjR/iDWm9GCnhNQ0FlqMWHizuoOxU+WIuCwDAtmG3IjxCvTbggsK/ly56zAR+45lxbI6iTd3AqX0ZcUzCY0pDQlOi1hprQCADgf7/+Ily7e0vjFWFH8WTiGgyLIa4qoY7SVIiZpx7igBuv4yzee2QeF582HHiO3EcQ5M8S/TFPzxYwkNRqki2Dj8trGgLs+kiA/Zs2iriKqQq2jaZw9sZsw+9qhNQIOpBKFI2gzeGj7YILgkZF0rh6T1T/4com6tu7OwFuXhB/NyOCRvCtfVPYNprChU6oYxDiPpfDBIFzLT99/R4QEdJxzXVaBnFqWcd4Ju5mtPpLL4dhMkQKH10pvIl7mFkI8GoEUQvahUUNxdSqs/ivvvUoHn92Ga99brDZs17UkBihVayY2DmUiaSxe/IIkt56Q6WK5Sk4F8b3/+ilaEfJsc4cTfoc8WYLdha3P3y0XWQd01CuzowUqGo62YQWGD7KH66JgUTHJ5wZARqBGSF89C+++Sg+2yDpLIppqGR4Y87TDUxDs/mym1U8OZR0I3UaYVkMq3nb8Vl01CSqKJ3P/OvFfP4Cw7Iwl9fxyR8+hV+5eAve9qLgqDmuSYg1sjj+JLKxTDxSVJ94XIO+wnNRNALAFsztKD4pBUEHwh/+gaQWGj661s7iqPCQtkJDjcAEkf3wBw30/IGYyCYCH75OgptvRN9OxbQaCoK8bnhMZEEYUUxDFcszqKQbmIZmc7qbQ7BxMOlG6jSiWdNQs/zy7s0A6msdXBAoFL3efriPwG5Kz82Yl54+Fnp+ikKIq0pD0xAAjGXjruCoZyISB3quBfHfuBxRI2gX0jTUgfCBZSgVCw0frde4fj3JCsXD6lGqmG499mCNwH64xgfidc0cnQAf8Jt1FpcqpiexLghRuIRpWXaoYXVQSce1uj6aU7kyztlkR/VMDiXx48OzdY+BYzbhLF4J77nybLzwjHGPH8APn2hENQsBPo1AEU1DiqcceqN9xjUl2Fns2PV5r+axbML1Dw2nY8iVDagK1UwMRNMQF1Z8YlCsRNMI2kVnTiv7HH5jDiZja5ZZ3C5cH0EjZ7Ez44mpSqizOB1XkY5rkTtorRfc4egPH2V12htWTLveTyNtx4gUNWR6Zo/1TEOMMUcjsE1DG4eSWHLaSjbCWmWNAACev3M0MI6fwzWCZgZJvq6qeNtmaiqhYlbr/zQSBGIpdpGlUgWpmOrm1oxn4ohp9vfwAnwjASGeooByBUGlWmySR+CtBVIQdCB8hjycjiFfNmp6ylYsq2NNQ7x6ab1qmYCjEThtAAMdcKWK2xSn830Etc5iPvsL0wq4gA/KVBURheQ/fu8QvnPg2Zp1yoZV09LQEGa6IstluxY+Nw3xInVRCgU2U2JiteATjWY0Ai4k/Q5czcks5tepkc9BLLwoslS0ewYPOMc2PpBwfQSjGdtMtGW42qiJX0IxgokfI/ed5ctGTbfA1aSl0YSIRonoDiI66PwPbHRERJ8homki2u9b/mEiOk5EDzl/1wRt329wc8DWkRQMi7mdwTjtbkzTTqJGDR1fKNq21FDTkP1whanjqwljDF/eezRSE3gg2FnMI4nC/ATc5NfIR8C1DV7S41/vfapmnbKvHAE3rQSZFU+5bUarPgIgmiCwrNU1DUWBn1vU0FGgqhH4HbiaYjuL+fMWxTTkzyyumBaemStgMKW59/5YJuEe30BSw1ff8UJc/8Id7jbD6XhNQbm4qoDIFgSGo6VkIoSgtotWp5XvA3AXY+wsAHc574P4LICrQz67iTG22/m7PWSdvoIPKLzUs/8htUtMdKZGkIrZiUH1BIFuWPjpM/PYs30UcUc997NYrDhx7uqaawRHZgv446/sw7f2TUVav2LVOosNi9eiDxYE3HTTUCNwsq4///YX4FXnb8KpXG2op60ReE1DQLB5jlfHHHejhuzs4ij1oQxr/Scg3FwSNZlMXNcvPGzTEHN/g0bCRWzOxPmDLz+M+w7P4kVnTrj+sbFsNWoorim4YOuQxzQ0nIrVaB9EhKSmolQx3ezisJLwq0Gro8m1AG5xXt8C4DVBKzHGfgBgrsXv6hv4wMgbcUz5BYHV3n4E7URRCOmY6umJ6+eR4wsoVSxcevqo4yMINw3xWZjfPLaa8OqpUePrA8NHHZNOWImFoltKoL4gMCwLCtn27YmBRGCZYztqSHAWu0l9tb/BEae7F68e20xXOTNiGerVxHUWN6UReBvXc2Iq1wh468v6A6/YnInzyLEFvGLXRnzo1eci60T+jAuCgH+3qG0Mp2OB38XLZ3B/TTdpBBsZY1MA4PzfsIJ9vIuI9jnmo9Ae2kR0AxHtJaK9MzMzKz3eroAPKFVB4A3vM6zOdRYD9g1cTyO4/yl7TvC8HaP2QB8SPjokzJzW0jzEB9t6TUZEKgE+Ai4AwpzFxYp9fRrV+dHNqj9oIpvAYrFSMxjVOIudQSbIAXz4VB5xVcFmp29BKq5iMKlFMw1FbEyzmlSdxdGHLiJCXFNqnhm/j6CRcImrtaahXNnAxEACRFTVCDIJ11nMBbR4vKOZROBsP6kptkZQ7kBBQER3EtH+gL9r2/D9HwdwBoDdAKYAfDRsRcbYzYyxPYyxPRMTE2346s6FDyiTQ0nEVKrRCCodbBoC7BDSeuGjP316HmdMZDCWtZ1qQa0ql4oVDAqCIKjq42rBBUGjBvCcoPBRbhoK8xEUomoEJnMHKF422m8e8juL0wkuCGqFzFMzeWwfS3ucvuMDCbciaT1WO48gClwjaDa0Mun0MhDRnKJzQY3ig0jEap3FyyXD1QQ2DiYxkNBs35bq1QREjeD3rzoLf33dBbXHGFNRrJhuoIW/bexq0lDkMMauDPuMiE4S0SRjbIqIJgFMN/PljLGTwr4+CeCbzWzfq/CBMRFTnYQfv4+gc01DgD2TqZfZulQysGHANknEA0xDlsWwXDY8gqBcsYBkza5WBd5sfKEQ0VkcED4a1VncKHy0YlruTJYLgpnlMh45toiP330In3v7C+za9Z7wUc3zHSKHT+Vxuq+7XCqmRqpAaq1yiYko8Jl0M1FDgP0sBTqLxfDRCD4CsZwE70HMo4XeevkOvPrCSSgKVX0EromoOqjvmhwMzAZOxGzTEPftpLslagjAbQCud15fD+DWZjZ2hAfnOgD7w9btJ0RVdXIoiRMLftNQ54aPAtWaMWFUTMtN9Q8yDS2XDbeBeb068KvFSk1D/BjFDnONwkdLlfr+D7s/tVcjmFku45v7TuDhY4t4/9ceiewsNi2Gp2fzOH3CW6QsGVMjmd7MVS4xEQXuLG5WECRjimuu4bjO4ojho/7MYn6Pc5NQOq65AR6xEI0gXqf9ZMrxQXCNIErhunbR6mhyI4CriOgggKuc9yCizUTkRgAR0RcB3AfgHCI6RkRvdz76CBE9QkT7ALwUwO+3eDw9AZ8hx1TCpqGUx5HHGLP7EXRo+Chg38D1fAQVITM6FhA1xE0yto+Ad4ZauzITM7nmBIHfWWwIFS0baQRAfSEnXitREPAOXt/aN4WpxZLXNOQIgnd/6SG850s/c5cfny+iYrIajSDh2KYb0QnO4pRrGmpSI9BqNYIYDx91BUFzzuKc03EsqJsa19j9zuJEnQlcMqaiqJuubye9hgllLYkcxtgsgJcHLD8B4Brh/ZtCtn9zK9/fq1RMy62lsnkoie8cKIExBqJqmnon+wgamYYqRnWWGxQ1xAfgQadZB7A+GkFU0xAPH62YtX6BRj4CoNbGL2II2hNvNj+zXK4RtMkA05BpMTw6teQuP3zK7ha2cyLj21aNlDPRCc7izApNQ8mYUlOxV1N9zuJImcViNVinQm5If2Xxf1D0UO0x2r9D1UfQPRqBZBXQDcsdKDcNJaEbdoVEoGp77uyooeZMQ36zBB+UeGYx0Djevp006yz2awRiNnCoRiDMwOvNxkXtL64pGE7HcCpXrrm+QRqBfQ7V9fh5bRzwOluSMaVhzSOgs5zFzYSPAnbEFU+i49jho9UyH5ESykTTENcIAkw4YaahepoM/x2qUUNdohFIVgfdtNwbnZcAmFosYSybcBtgD9ap2b7eZOIaciUDh2dyNfZowD4/rjoHOYtF0xA3s6yVRmA6mdwKwW252cgfYwg5A5av/HRYQploGqon5OxrVf3+iaydS5ArG9g8lMQJJ5BAHGDE1+JMf9k1ZXjvHZ7I1AiLrb9pKK4p0BRqKqEMAG56w24QvMeuKgTDtE1DmkINhVxC8/pSXB9BwLMY8zmJo2oEdkJZ9zmLJatAxbTcG8bN/Fz0JjmFNY/vBDIJu3z2yz76fRwOaF5eEQRdUNE5PovlmcXA2mkE8wUdpsXchKulUuNibGLSmG5ankJxYQllftNQ6L79gmAggenlEnIlw3VMAt7QR9EZWdBNV9ByQeAfuHi0SiNsZ/H6a6Jnbsi6v09UhtNxDKW9tnxNJVScmkxRTE3+onPu9QzUCKpaHICacNIguCAo6HZeyFpqX1IQdCCiaaiqEdiRQ9xENNqGhtWrhegs5T1cRcRImLim1DRxWRQ0gmrU0No4i7n55IwNWc+x1EMc+HXT8gz+7TANiaHCo5k45gsV5HUDGwYT7vJ6jk6uYS07VTL9TtNkQMZsEEYHmIYA4L/e8xL85otPb3k/MSF8NIrz2a6LxdwkQd4xLshZHPeFj/KktrqCQLMFcq5srGnEECAFQUdSMZkb6jaWTUBTqkll8/nO1wiuPm8S52y0690HlUSoCIKO/xfNQ0ulChSyTUxuQtkaZRZzdX/riK2JRREEFVEjMLwNaayQwy4KoZ31NIKKzzQ1nI5hoaAjV7IHC+4/qNfEhGs1y6XgNpDJWG0NnSDWogz1WqKpBIvZgjiaRuD0LXbu1VyIqQ0QfALC75JQlboC2/YRmCiUjTU1CwFSEHQkoo9AVciTVDbnmIZGO1gQXLB1CF95x2UAqslZIhXLElLwveV3gWpWsaKQ+0CtVYkJLpB4qOZChHpDHo3AsHylJoKP22MaqqsRWB7H6Eg6jsViBcvOrJGbefwDzK/u2YpzJwcBVIXZcrkSOGglnHpOjTqqdUIZ6nbCJyEFPaog8AYu5MoV218RsO1F24bx3qufg8tOH3OXxQOym0VSMRWGxdxeHGuJFAQdiGgaAmzz0AnHNMQ1guF0rTraSWQTGpIxJVgjEMom8IFsWbDFLxYrbjPvta41xP0VvDpnJNNQPY2gTkIZH1Pr+gh8BQaH03FYzP6ebEIL7fP7kdddhL+49jwAomnICDRj8KYo9cxDPH+lkzPam4ULtbxuRCpZ4U5KHDNlzikvEZQgFlMVvOOKMzxtMhONTEPOunN5XZqGJF5nMWCHkLoaQb6CbEKLdOOuJ0SE8WyiRiPg/gCeB8Gjn8RwyKWS3YsAwNprBM73TDiCIEoIacXnI/CGjwZvU9RNDDt+nnoDsG54TUMjwgQgk9DcgT0oioY3ROeRQ0thpiFXKwu/xgXdhGmxQEHSrXCzWqFsRgpH9WsEy6XmbPkNfQTObzib190KsmuFFAQdiB01UB3oNw+nMLVoJ5XNF3SMZLrjYZwYSLhZuhw3a9oxDWUT9rnUaAQpr8ljrXwE3JTDNYK5fBRncbhGUM80xLW6egOwYbEa0xAnm9DcGHa/AxiAK0wXBWfxYB2NoJ7Tuhq23B33XhT4NcvrRjTTkKs5OYKgHKxhhfHK8zbhF84OL5jJ9z+b09e04Bwg8wg6knzZcDtHAXbN+LJhYb5QwVxe7+iIIZHxbALPzBY8y7ggEDs4AfYgxVkqVrBhwI7aqZqG1iZqiBf8S8UVjDjJW43wh4+KhoIwZ3GpYjoO/3zdcxOLzgFek2A2qbnXL0iY8EGbh+PWcxYD9bUurlUEbd+t8Ota0E1sGGh8Xvye5ZOSXMkITCYL40+uObfu525nuYq5piWoAakRdCR24+rqjcCFwvRyydEIukMQcI1ALKrGzSYxnyAQTUO8FwFQ+/CtNtw0FFOV0EYwfvzOYk/4aJ0OZdzMUz+PgHlm+8PCJCCTqBY5C/LhJmMKYiq5g/hyqeKai/zrAY00gmrZj16Bl5zIlaNqBN5JSa5sBCaTrRTRCrDWGoEUBB2IHUdcvRH4zZYvm12nEczldez8k9vxw4N2M6FqQT2vs1hM3FoSBiy7pC+tmY+Am3I0RxBML0dr4cjRDctjKjJDVIJipeojqDcAi1nYgNdHkE1o+OOrz8FfXXc+XnpObU8oIsJQKobFYgUV00KpYgXOYLn5rd5xLNUJlexWVNdHEM1Z7A9caHe8vxgCzKPW1gopCDqQfNnwFJzis4N82cB8vrs0As4PD54CUJ3Z88GNmy/4jLNsmChVLFcjALCmfYt1V2Mhu5xDBNOQ6Cyu+MIwg5zFjDEUdMMNAa4Xw+/PLB5MxtzZvx2ZpeLXXrA9tPTDYDKGpWIltLwEUJ3p1vNVcKd5J5c2aRZuGsrrZqSEMjfL3ag6i9tpwhE1grOdPJy1QgqCDsOyGAq66Yka4Mkl8wUded3s6BwCEVFz4TNR10cgFOGKqeQm5xQCujOFtbNcDbhpKC6Yhhr1S/Y7iyshZaj3HVvABX/2Hdz9xAwqJsPGwSRUhXDbwyfwhfufDj4ek3kqzSoKuUIyilliIBXDUslwBW298NFSHV9FLzuLgWjVTNPChAyobRHaKklBKzlnkxQEfQ0vOCWahngVQl6uodNzCDjbx9Lua26n9vsIeK9XPtAUHPOEmFkZV6OVQGgHftMQT/lvtA0fEHTTchvXA1VBYFoM7/i/P8Vy2cA3HjoBANgwkEBCU3BwOod/+f7hwH3rplXTUIVHDmUjZJ8OJjU7Aa2ORpCM0PNhqY4g6VbEnh5RNAJevXTW0RLrlQ9fCal49Ri2jaTrrNl+pCDoMNxa5IJGwF/zekPDqe7QCM7fMoS9H7wSmwaTbuSK30cA2IMLn7HyqpxJQSNIxNZQIxBNQ0IjmHoYFnNNeWXD8jamcbSJvUfmcNzpNHdw2i7Et3Ew6Q5GvIZUzb5Nq6aOPp8IRClTPJSKYWap5IaQBmsEjXM1lksGYiq1dQa83jSrEYym4yACZnI6GLOL1TXbIKceolBZ6yqvvfOr9gj+9ndAtUEFTyrrJofdeDaBwZQmaAReHwHgbXbPBUFKsJfaGsEa+Qi4D0NRMJHl0VoNBIHJkIpX8x0MT60h+7WoVTzmNIvZMJBwnbC5slGj9ZgWg8VqcwRG0nEkY0qkdqUve84GnFgs4aY7ngAQohFEyiOoYCAZC22z2I2ImnWUAV1TFYym45jNld16Q802yKmH6CNYa1o6CyIaJaI7iOig838kYJ1tRPQ9InqMiA4Q0bub2b7fcJtSCGp/MqaACG7huW4SBIDjsAwxDQH2+fABseiahtZHIzAsC6pCUBRyq3s20ggqpuX+Xn5nMRcK/PizCc1dJlYPBWq1Ai40/U2IRjPxyLb66y7egqvP24S9T88DCLbxJyJkFi8Vg3MQuplto1XzS9QBfSwbx6lcOXKf42bg1/etl+9o2z6j0upZvA/AXYyxswDc5bz3YwD4Q8bYuQAuBfBOItrVxPZ9RbU7UfWhIyJk4hpOOr2Luy2WezAVi2Aa8gqC5DppBGI9HV5mIoppiPeXtYvO1WoEfAZ5mjP4ZBNaTYXJ2VxVEOTLBm74/IMAartxvfOlZ+KmN+yOdD5EhL/91YvcgW44ICs9qkbQS45iABjLxIV+AdFm4+PZBGZzutDnuL0awcN/+gp86Bd3NV65zbR6FtcCuMV5fQuA1/hXYIxNMcZ+6rxeBvAYgC1Rt+838o5pxB+fnEmoOOUMFN02MxtMVk1DeoBpaCCpIVfmPgJbIKQ8xbqaCx89OlfAjw6dWtGxVoRwzaFUDDGVIpiGqhqBXWKi1kfAj3/HuC0INgTEic8LlU6fOLmMHzwx49kHZ8d4BpefOR75nLIJDQ/96VX4z999YYMSE/Uyi3tPIyAibHS0sqgdz8ac+llVjaC95pyhdGxdusC1Kgg2MsamAHvAB1Cb1SJARDsAXAzg/ma3J6IbiGgvEe2dmZlp8bA7F64RpH2OQNFU1G0zM1sjcExDRpBGoNVoBGlf+GgzUUOf/OFhvPvfH1rRsYqCwA7VjDesQFoxGRKabb7zF50zajQCOxOYO6L//YZL8b9fdyEAr2lI3MdwGzTAdFzDc08LtryqTtJe/fDR3tMIgGr/5qg9kMezcZzK6W6EVTt9BOtJQxFPRHcC2BTw0Qea+SIiygL4KoD3MMaWmtkWABhjNwO4GQD27NlTP7C7iwlyFgNVwRBTg+ufdzK2j8BwSxkD3geICwLGGIq6PWCmRB9BQIP7euTLJkr6ysJNK4a31HImobrCOQzDsusBxdTauv6Wz0fATUO8bMgLTh9zk4dE0xBf/yOvuxDXXbwFq01Sq9+cphd9BED1d4g6oI9nE8iVDden1W3PYhgNf1nG2JVhnxHRSSKaZIxNEdEkgOmQ9WKwhcAXGGP/KXwUaft+IshHAFTj6ge7MHJjMKXBdBLlgnwE2UQMpsVQqlgocNOQIAgGUzGP2UTkiw88g6t2bXSrhQJ2os9Km91XLG8mbyauuccUhmExaKqCVExFSTcDW1Xy8+a5FaJpaCgVg6qQRyPgGtA5GwcCK4u2m0RMbagR9FIOAYc77Bv9xpxxJ5dgygkFjmpS6nRaPYvbAFzvvL4ewK3+FcgetT4N4DHG2Mea3b7fcE1DvlAyriF046zMrYJZqoT6CAB7sOEOS9FHsHM8g5NL5ZrErlO5Mv7kPx/BrU6CVqliQjesmi5hzVDxFXnLJNTGCWUmQ0whZOIq8rrpKUJnWn4fgW0a2jRUrS6rKISRdAyz+VqNYK1MD7xNYhCGaSGvm25p8F7CLei41LiUCFAtT85zQqI6mTudVu+yGwFcRUQHAVzlvAcRbSai2511LgfwZgAvI6KHnL9r6m3fz+TKJjJxtcZhxG3m3TgrcxukFA23HEPc5yMA7KqjxYoJTSHPYHzGhD14PjWT9+yXl6PgyWjP+dB/4bUf/xHKhgWLhTeOXyxWQn0Odj9l0TSkedpKBmE4fYXTCQ35shFYfZQP7JuHkvjkW/bg9Xu2efYxmoljLl8djFYjTr0eCU0JNQ3xJMe17pq1FrzwDLuV5EXbhiOtP+YTBL2iEbT0yzLGZgG8PGD5CQDXOK/vARBoywjbvp/Jl4MLWXFncbdrBEGmIR6meSqno6CbHrMQAJw+YfcmOHwqhwu2DrnLC5VqnX1uVnnk+KLbJ7ZiWlCV2hnbRX/+3/iFsydwy9ueX/OZEWAaemauULOeSMVpJ5lJaMg7nbw4rkbgtOckIly1a2PNPmxBIJqGvH0bVptkTA3VCLjQXM+Ep9Xiwq3DeOhPr/KU964HNw2dcDWC3hAEvXEWPURODy5ty53FXSkIHJPCUlEQBMJM1y3lkCujVDE9ZiHAtqsrBDw5k4dpMRw4sQigmoWcKxm4RwgX5QNXkJ+A14n5/hPBkWe6yTwZu5mE6moeIkulCj533xEUdMPWCBQFmbiKQtkINQ3Vm92PZRKBpqG1mnEm6/gIViN5qpOIKgQAYMDpqDfvdK7rFY2gN86ihyiEaARcOHRjCF+Qj0As+LXBCeGbWS6jGKARJDQV20bTeHImh289MoVX/8M9eHax5AqC5XLFjbk/e2PWHbgqAZFGT5y06/yE1cypGBbigmkoHdcCo4a+9/g0/vTWA7jsr7+L+UIFmqMR5BzTED+/qkZg1hUEEwMJPLtYcoWIO/iukQ1aLPznxz2WHtQImoXfm3NO8ILUCCSrQr5semLoOWnXNNSFgkDwEVSM2hITgykNcVXB9HLJNg0FDDinj2dweCaPI6fyYMx2FHPb/XLJwANPzQGwHbd8Nh2kETxxchkAsDWkuqPfNJRNaMjrRk0pah4+yHMMNMdZXHCihjSVoBBgCT6CmBoe7fX8naMo6CYePrbgrg+snY9gLBP3hK+KcA2rVwa9Vog7ZdMXHEHQK8Kx++wMPU6ubGBSiCjhZLrYNCQ6g227PbndoQA7w5PX/i9WajUCANg5nsWPD8+5FVhzZcNNPlsuGW4nsVLFdE07XOgAtkP3j/7jYdfJF9ZQRDeZpxxwOqHCYnbWrXhcOd/smTuLbVMRQ0xRYCrMdRxXTFZ3UH/hGWNQCLjj0WkAtPaCIBsPrYBaXmMzVaeTjmtYKNgTgF4Rjr1xFj3EUkhf2XQXO4tjqoJkTEG+bDiZu7Uz43EuCEI0gu1jaRQrJg6csHMRcyXDNQ3NLJfd8ghlJ3wU8GoEJ5fL+PpDJ/CTI3bxtWJI3LjfNMRNcrxPxBfufxpvuvnHrrmIC7SYQm4VVdOyoKq2sBMTyuoNGsPpOC7YOoxPfP9JvPbjP8LxhQI0n8BcTUYzCRQrZmA8/WrU1elmMnHVFfC9Ihx74yx6iKViJbAdIG9U020F5zjZRAzLZcPpwVt7201kqxpBkGls22gKAKqCoGy4gxbXErIJDaWK6ZoyxFwCw2cmygc4gAEnS1io/88FMB/4P/C1/bjv8CwWixUMJDWMOd3iNFVBOq6iVLFQNux9qESuj6BsWIg3qEvzCiGaaGa5vKblC/h5BJmHet1Z3Cz+rPdeoDfOokewLIblsuHp18upZhZ3n0YAOIXlSrZGEDQznhiwi3kVK2ZgmCIvzcAH1uVSBUVHC+DRmpNDSY9GIAoCvmzTYBIXbh0KzSStmMwT0cQFcL5surN7wC4JPpDQ3LhyTSU3xHepVIHmlLI2hMzieB0fAQD8zi+cgb94zfkAbDPaWgoC3v40yDzE6+q0u8BatyKaFaVpSNJWZpbLmC/oYCx41r9jLIN0XMUZTkx9t8HNJnYtn2BBMJvXsVwyAjUCv3N3uWzUmHc2DSVhWMz1HXgEgfP6w798Hi47fSw0ScxvuuICuKAbOHwq5y4/sVBEJlHVCGKK4g4Qi8UKVIWgKeRxFjca2FWFsHMs4+5jLQeZ0WwdQSA1Ag/8/oxrSteVewmjO6eXPYZhWnj5R+92s02DBMFpY2k8+j+vXutDaxvZhK0RDKdjNT14AVsQMGYLxCAfQTKmYsNAwi0JnSsZNaWpNw/Z5iM+cdcFZzEvdpfQFKTjmt1S0skIFqn4WkPywT1XNvDkjCAIFovYMZZx+9ja4aP2cS8V7baOqkJC+KgVqc0jd1QvFiuBTvPVggu0U7naUgtVH4HUCIBqcmeiR7QBQGoEHUGhYmKpZODhowsAEGga6naySQ3LZaOmlg9nQigalwwZAE8TOkotl6pRQ5xNvmirINNQTFXcAbsQkElrm4a81UcBoKCb+NkzC+7yhQL3ESTc/fMBgmsEiuAjaOQs5nCz2EJhbTUCbuL6H1/Zh+d86Nuez8prHMHU6XAB3SuOYkAKgo6AZ67yGWc3Jo01YiChYblUsWv5KLW33U6nGBsApGPBiipvLUjkhI8K5h2iagExjigIxF7JrrknwGFcMWtLTAD29z0zV/BoK5m4hvGBqkmFZ38vFiuIqQo0QSPw7zcMfmyGxRo6l9tJJq66A32pYqGom/jvA88CqOYRSNOQDb8nesU/AEhB0BFwx+W8E5vcqxpBjoePBpiGzt6YdUM1xTh+kR1jGRABW0dSWC4ZHjv/cCpW41sI8hHEtapGkA9wGNcIggQXGgamFkt4zuSA55zGM9U6SaKzmPc9NpvwEQDeqqtrOQMnIk9pky/vPYobPv8gnpktyDwCH1zg90oyGSAFQUfgd1z2Yrlf7iMICx8lIuzZYXfQCrNFv+Wy7fjMbzwPk0MpJ2rIBA+zH83Ea2zwutDlSzQN1dcIahvTALZGcHyhiOdsGvScEy+Cd8GWQVdoMAY3B8AbPtqcIFjrGbjoKH7qlF3pdalUqSa39dAMuBX4hKOXNKTeOZMuxi8IelEjGEjGYFgMyyUj1ETyvB2jABBa7XMkE8dLz9mAQUe7KOqmWx9+NBOvESBiraGKqBHEgzUCy2IwLa8PI+6YeI7OFaEbFs7akHWFTzah4dzJQdz7vpfhLZftcIUGYOcVqD7TUJSBtFNi1J+etQVBvmygbJh2FJQUBACq5rte8pn0zpl0MeKApJC3P3GvkHXyH+YLeuiA+Osv2I4XnzWONz3/tPr7ckJRCxXD7TA1ko7XmC6CfARxpxQEUNuVqmLVlsgmsovJHZy2axRtGUm5UV1cA9gynIKiVH0PgB0KqpIQPmpG0wh4tBE/1rXkLZdtd19zYVyomChXrJ6a/bZKRmoEktVAdHoOpmI1TWl6gQFn0JzL66HF14bSMXz+7S/AmRvq50pkneS0om66lUsDNQJREPBid4JG4NfEeIip//gycRUHp21H/uahlOvMz/qS+zLCbH7nWAaqQm4jnqhRQ0TkmofWesb5P689H9/6vRcBAI7O29nahbLd9rOXBr1WkRqBZFUQyxz3YsQQUK3ZU880FH1fMTt8VDcxko5jIKFhcihV10dQFqKGuPnF7yMwzFqNAACG0nG3RPPm4aTrw8kmvIJHNJ1ccc6EXWuoSWcxIIQnrsNAw82S3C9Q0A2UK9GPvR9wncU9lFfR0q9LRKNEdAcRHXT+jwSss42IvkdEjxHRASJ6t/DZh4noeEALy75CjIfvRf8A4J09t5ooNZDUoJsWFooVpOMqvv6uy/H2F+9soBFUTUPc9HZisehxkLq9EnyC4JrzN7mvRzNx9zfKJsJ/q8vPGnd9BJZlVyGNKgDXSyMAapMZC7pdu6mXBr1WkeGjtbwPwF2MsbMA3OW892MA+EPG2LkALgXwTiLaJXx+E2Nst/N3e8D2HcltD5/Atx+Zasu+RBNFzwoCITTx7I0DddZsDK/AWtBNt+xGNqHVaARhzmI+o/u7Ow/itz+/FwsFHZ+99yl8/O4n7XV8pqE3PL/aX5iIXK0tkwgfHAeTMShk1xpqtv/wegqCbFyDaJnM6wbKhjQNiaR7MKGsVa/ktQCucF7fAuBuAO8VV2CMTQGYcl4vE9FjALYAeLTF715X/uX7T0JTFbzqgsmW91UQTUM9GDoKeMtnn7d5sM6ajRGFiqhd+IvVhWUWi93RHp9axie+fxif+P6T7jL/zH3DQBI3vOR014nr+ggCehp88bcudfvactMQFwRRB1OeWR1fo+5kIopCGEjG3IY7Rd2Eblg9Nei1CvcR9JJwbHXU2egM9GCMTRHRhnorE9EOABcDuF9Y/C4ieguAvbA1h/mQbW8AcAMAnHZa/aiStWChUHETbVqFz2xVhTCUit4/tZsQB81d7RQEdeLuywFRQ5pCnkJhy2UDPz4869kuKEzy/dec676u+ghqH5/LzhhzX3PTUKXJEg3p2PrOOAdTmisI8mXTzoHoITNIq3CNspf8Jg3PhIjuJKL9AX/XNvNFRJQF8FUA72GMLTmLPw7gDAC7YWsNHw3bnjF2M2NsD2Nsz8TERDNfvSosFis45TRbb5W8biId13DTr+7Gb754ZxuOrvMQfQQ80meliFFF6XoagVB0Tnc6hAVVi3z42AJ2bxt23zcqF131EdSfR/F+BHqIEzqMlKsRrM9AI5onixVD+gh8uEXneuiaNLzTGGNXMsbOD/i7FcBJIpoEAOf/dNA+iCgGWwh8gTH2n8K+TzLGTMaYBeCTAJ7fjpNabSqmhZxjzjnhtD4M4pYfHcEhJ/68HkXdLr185a6NXVtmuhHtfGhOn8jidKc2kTh71xTy2Lf9pqGwgZUx4EVnjrvvGw3Yl50xjivP3ejW8A+DawTNZuaup48A8EaucY1AmoaqpOJ9qBE04DYA1zuvrwdwq38FsqdgnwbwGGPsY77PRAP7dQD2t3g8TcMYw9s/+xN89cFjkbdZctRmAG4PXD+WxfBntx3AF+5/puH+8npwV65e44pzJvDhX9rVeMUI/N7LzwLg1QiIyNUKsgmtJqFMzA+47uIteN0lW13Bcc6mqgO7UQbtJdtH8Knr9zRcT1UIJmu+Ef16ho8CXkFQ0O1y371kD2+VTFxFTKWGGmE30eqZ3Ajgy0T0dgDPAHg9ABDRZgCfYoxdA+ByAG8G8AgRPeRs934nQugjRLQbAANwBMBvt3g8TXNsvoi7Hp9G2bDw2ku2RtpmURAEYRoBNwccnw/XGA6cWMQHv74f00tlbBxMhK7XK3z2re1T+F5z8RacuSGLcye9/oaEpqCgm8gkVE/P4oovs/emN+wGANz/1CyOzhVxplM6wmK1CWUrxdYIrOZNQ7H1FQTcNBRXFSd8tHGbzX5CUxX8229dirMaJD52Ey0JAsbYLICXByw/AeAa5/U9AAKfLMbYm1v5/nbwwFNzAICfPTMf2KgkiAVRIwgZ6Lkj+VjI57O5Mn7x/9zjvhfLMEuicf6WoZplyZgKogrScc3NFAYQWuxux1gGx+eL2DmewcbBJKYWSy0nvHHsfgTNN39fb9MDd4ZPDieR102UK6bUCHzwuli9Qt//uj85YguCvG7i8Wcb2/MBYLFQFQQPHVv0dK7i6K4gCC6g9qMnvZEq/WAaWgsSmoKEpiCuKp48gjAfwYvOHMeLz5pAMqa6jW1MoTdxK2gKwRJ9BF2QRwBUNYLNQykUdUOWmOgD+vbXLVVM3PC5vfjOgWdd88JeRyg0gpuGVIXwgydm8Kq//2HNOryZx1LJwFKpUvM5z2jl9WmkIGgPyZiKuKogplGAj6D2dv/tXzgDt7zNNlld4GgYhtkeQaAqBMOyXM2keR/B+twT1+7egg+9ehc2D6dsZ7EsMdHz9O2v++RMDv/96EnMFyp40/O3YfNQEg8KrQjrsVCwB/EbXnI6AHu2yZh38BD76QaZj7gguHDrMAC4FTElrZHQFCRiKmKq4vMRsIaD2fuvORcfee2FuPzMsbrrRUVRCBYDdNOeFDRdYmKdwke3jabx9hftRDqu2rWGDKunQiUltfStIJjP27P0T/z6Jfi1F2zH+VuGcODEYqRtuY/gD686G++9+jkAUJNcJg5CQX6C+YKO4XQMOyds30C6h7odrScJrhGoSk34aCMncDKm4lefty0w12AlaF0aPspJJ1TkZfXRvqBvf905Z1Z/5ga7XPCuzYN46lS+pkZ9EIvFCgYSGjRVcStQ8uqUnHJFFAQF3PrQcXz+x09Xvz+vYzQdx7YRuw9vm8zSfU8ypiIRc3wEEZzFqwlvXt9s8/f1dhZz0jGtWh5D5hH0NH376847ppmRtJ0UdN7mITAGPDbV2GG8WKi4VRp5xmyu7BUEokZwfL6Id3/pIXzo69U0ibm8jpFMHBsGeM/bcgtnI+FsGEhgIptATK31Eaz1wKoqtuP5qNPkZSQdraAgr8u03nHqYlE9WWKit+nbX3cur4OoGiHB6988OrVUbzMAtkYwnPaWIs6X62kEwT6CkXQc444gEENSJSvnz35pF/7lzZfYPoIIUUOriaooMBnDt/c/i4tPG8ZYNlquyIvPmsAnfv25LRfnaxWx41ovNWqX1NK3gmC+oGMoFXPzBjYPJTGUiuHRCH6CBY8gqDZcEeEOwpF0LLAH73xBx1gmjktPH8Vrdm/Gn7763Jp1JM0zkIxhOB1HTFMiRQ2tJqoCzCyXceDEEq45P3qVWlUhXH3+ZNt8FSsl3SH9kyWrT9+GqnAbPYeIcN7mQTxyPIIgKOhuSYKBENMQ1wjOmMjioaMLns8YY65pKKGp+Ls3XtzKqUgC8PsIokQNtRsxDPVqoblNtyAFQf/Qt7/ufMEeiEUu2jaMx6eWG1YUnS9UMOwIEd7APFf2mna4j+D0iQwMnyc4VzZQMRlGM73ZhKYT8PsI7Kihtb3deZXUj7z2QmwbTa/pd7eDDYN2gt2W4RSuOKduhXlJl9PHGkEFW4ZTnmW7tw3DsBj2H1/EnpAUcsO0MF/QMe7Ye7lpKBcSNbRz3FuPhDHmhq6OZnq/vtB6EfeZhnTTQlxbW1PL2y7fibddvhOKsr4mnpWye9swvvOel+DMDVm3KY+kN+lfjSCv18zIL3Zq0vtNOSJzBR2MARNOF6qqaaiqRVgWc5uinDHhrSFUMZkbuio1gtUjpiqe3I718BEoCnWtEOCcs2lACoE+oO8EwWKxgk/+4DCml0s1pqENg0lsHkriZ3UEwWzOHsR5BEhCs1sfctPQvYdO4fT33+6Wqzjd11+gbJiYy9uhoiPp3uxG1gnEfQllFdllSyIJpe+ejP/YexR/dftjsBg8zmLOc7eP4P7Ds55BRITH+3PTEBEhk9CQKxko6iZ+7VN2F859x2yn85bhlMdJWapYmHNNQ1IQrBaxoIQy6fCUSALpuydDbKLu1wgA4FeeuwWncjr++8DJwO2rGkF122xCq+l9y6OIEpqCrYIvolQxcXB6GTGVsHGwtZaNknCSMQWmxZAvG2CMoWKyNTcNSSTdQt89GbowS9QCbJ+/cPYGbB1J4fM/PhK4vV8jAGzhkisZWCjq7rL5vI6YatuI3/uq5+ANe7YBsGsSPXx0AbsmB2t67Erax8WnjQCwy31zzUCGQEokwbT0ZBDRKBHdQUQHnf8jAeskiegBInqYiA4Q0Z83s327KTuhoa88byNe9pzakDhVIbzq/E148On5moqiAHAqpyOuKhgUNItsQkNeN7BUrEYOGRZzKza+8rxNeNm59ncVdAOPHFvERUKzdEn7ed6OUWTiKr7382mhQ5h0ekokQbQ6RXofgLsYY2cBuMt576cM4GWMsYsA7AZwNRFd2sT2bYVHkvyfN13s5gL4GcsmUDEZigH5BKdyZYxl456sT+4j4L2MuVNS9A3w2eiBE0vI6yZ2S0GwqsQ1BS86axx3Pz7tlpqQpiGJJJhWn4xrAdzivL4FwGv8KzAb3sIr5vzxqXbD7dsN1wjqRZDw+kMLhdr6P7OOIBDJJm0fwVKpglRMxZBTfkI0RXAz0P2OH0FqBKvP5WeO48RiyS36tt7VPCWSTqXVJ2MjY2wKAJz/gemHRKQ6jeunAdzBGLu/me3bid1kQ6lbx2XYEQS8E1m+bOCNN9+Hx59dwqmcjjFfItiAqxEYGExpbpJZPEAQPP7sMjSFsHNM9ihebbjGx/M2pEYgkQTTMLOYiO4EEFQo5QNRv4QxZgLYTUTDAL5GROczxvY32Mx/HDcAuAEATjvttGY2BWAneRUqJsqG1dBJy2f0XCPYd2wRPz48h/d+9RHM5so4e+OAZ/1sQkPO0QgGkzG3dnsiwDR0KlfGQFLr+kSjboA3++E9pmUegUQSTENBwBi7MuwzIjpJRJOMsSkimoQ946+3rwUiuhvA1QD2A4i8PWPsZgA3A8CePXuabuPyG5/9CX7wxAze9PxtDaNHhlP2THLRiQLimZUHji+CCNgw6NUIskkNBd3EXF7HYCrmrh+kEczldWwd6b66M91I2qmnzzU7qRFIJMG0+mTcBuB65/X1AG71r0BEE44mACJKAbgSwONRt28HxxeK+METMwCAgm427LY07NMI8k7XMsOy49EvPd3b05ZnGT91Ko/BpGAaUms1AosBg6m+LfG0pvB6+vx3lD4CiSSYVp+MGwFcRUQHAVzlvAcRbSai2511JgF8j4j2AfgJbB/BN+tt327+8buH3NcLhUrDRtyus9iZSRaEOkLpuIoX7PQWpJt0EsOml8sYTMXciqTi94jmqIGErDG0FmScMsrTyyX7fULmbUgkQbQ0NWWMzQJ4ecDyEwCucV7vAxBYcD9s+3bzixdM4v6nZnF4Jo+FYqWhaSgdVxFTyTUpiH2MLz9zvMbHsGmomiE8mIy55Sm8pqHqazG7WbJ6pFxBYCcBSgEskQTTF7ryi84axzuvOBOA3VSmkSAgIgyl4q5JoaBXu429/pKtNet7BEFKEzQC0TQkaARJOSCtBRnHNMQFQVYKYIkkkL55MvjscD6vY/NQqsHatp+AO4u5j+C+P3l5YMTRaDqOuKpANy0MJmNQFSdXQRAEqkJOsxQmNYI1gv/mM0vSNCSR1KMvNAKgOigslYyGzmLAziVwNYKyCVWhUE1CUQgbh2yH8WAqhqwz4Ph9Efz9oBQEa0JCU6AqJE1DEkkD+kYQpGPN9V8dTsc8UUPpuFo3CW2T4zAeTMbcaBV/lAr3Ewym5IC0FhAR0jEVhsWgKuTx00gkkip982SkPI24G5sIBlOxqrO4bHoaeQexyTE3DQjho36Bw79XmobWDv67ZxNaXUEukfQzfSMI0vEmNYJUHPMFHculCgoV03U8hjHpOIy94aM+QeDMSKWzeO3gvwUXzhKJpJa+EQQpYSCP5CNIx1DQTVz05/+NZ2bzbpZqGBtd05Dmxq/XmIakRrDmpGLymkskjegfQSD4CJIRTEPXXLAJW0dSsBjw2LPLrt0/jFeetxG/8cIdOG00LTWCDoJHCmWkRiCRhNI3gsBjGoqgEZy5YQCf+PVLAAC6Ybmz/DC2jqTx4V8+D5qquINOmEYgo4bWDq4JStOQRBJO3wgCu/Q0fx0tnnxioFpcLt3EQJINKDEBSI1gPeACXCaTSSTh9I0gICLXPBS1d+2o0Nw+3UR/4Y2DCXz4l3bhmgsmPculj2Dt4VFDA1IjkEhC6RtBAFTNQ1EFQUxVMOJUIm3GxkxE+I3Ld3o0CsDOI4irimxav4ZkpGlIImlIXwkCPjtMNDEQ88G8UR5BFDIJzS1xLVkb0tI0JJE0pK+ejmZNQwAwnk3giZO5tkSdvOOKM3DdxVta3o8kOmmpEUgkDemrp4NHkER1FgO2IADaoxFsHUnL7mRrTFrILJZIJMH0lWkovUKNAGiPIJCsPTwRUJqGJJJw+koQVH0E0U+76iOQA0k3IjUCiaQxLQkCIholojuI6KDzfyRgnSQRPUBEDxPRASL6c+GzDxPRcSJ6yPm7ppXjaYQrCJoyDdkhpLKWfXfCBbgM2ZVIwmlVI3gfgLsYY2cBuMt576cM4GWMsYsA7AZwNRFdKnx+E2Nst/N3e8D2bYObhpopRzzpVBUdkqWju5IXnTmO91x5Fi7cOrzehyKRdCytCoJrAdzivL4FwGv8KzCbnPM25vyxFr93RaxEI3jhGWP4lzdfgueeVqPsSLqATELDe648GzG1r6ygEklTtPp0bGSMTQGA839D0EpEpBLRQwCmAdzBGLtf+PhdRLSPiD4TZFoS9nEDEe0lor0zMzMrOthUkwllgN197JXnbZK17CUSSc/ScEQkojuJaH/A37VRv4QxZjLGdgPYCuD5RHS+89HHAZwB22Q0BeCjdfZxM2NsD2Nsz8TERNSv9pCOOeGjslOVRCKRuDT0oDHGrgz7jIhOEtEkY2yKiCZhz/jr7WuBiO4GcDWA/Yyxk8K+Pgngm5GPfAWk4rYAiEszgUQikbi0OiLeBuB65/X1AG71r0BEE0Q07LxOAbgSwOPOe7Eq23UA9rd4PHV51fmT+IOrzvYUk5NIJJJ+p9WYuhsBfJmI3g7gGQCvBwAi2gzgU4yxawBMAriFiFTYgufLjDE+8/8IEe2G7Tw+AuC3WzyeumwbTeP3Xn7Wan6FRCKRdB0tCQLG2CyAlwcsPwHgGuf1PgAXh2z/5la+XyKRSCStI43lEolE0udIQSCRSCR9jhQEEolE0udIQSCRSCR9jhQEEolE0udIQSCRSCR9jhQEEolE0ucQY+tSCLQliGgRwMEmNxsCsNih2wDAOIBTa/BdnbwNIK8DsLJrsNLv6uRt1upeWOl23XgdtjPGaou1Mca67g/Azb20jbPd3k49Pnkd1nybpq9BF5zTmlyHFu67vr4O3Woa+kaPbbNSOvmc5HVY+TYrpZPPaa2uw0q/p6+vQ1eahnoRItrLGNuz3sex3sjrIK8BR14Hm7W4Dt2qEfQiN6/3AXQI8jrIa8CR18Fm1a+D1AgkEomkz5EagUQikfQ5UhBIJBJJnyMFwSpBRJ8homki2i8su4iI7iOiR4joG0Q06CyPE9G/OssfJqIrhG3eQET7iOgAEX1k7c+kNYhoGxF9j4gec87h3c7yUSK6g4gOOv9HhG3+hIgOEdHPieiVAfu8TbyunU47r0E33w/NXgciGnPWzxHRP4bss6vuBaC916Ft98NKYm7lX6Q43pcAeC7s3sx82U8A/ILz+m0A/sJ5/U4A/+q83gDgQdhCegx257cJ57NbALx8vc+tyeswCeC5zusBAE8A2AXgIwDe5yx/H4C/cV7vAvAwgASAnQCeBKAK+/sVAP8mXtdO/2vXNej2+2EF1yED4EUAfgfAPwbsr+vuhXZeh3beD1IjWCUYYz8AMOdbfA6AHziv7wDwWuf1LgB3OdtNA1gAsAfA6QCeYIzNOOvdKWzTFTDGphhjP3VeLwN4DMAWANfCvnHh/H+N8/paAF9ijJUZY08BOATg+QBARFkAfwDgL9fsBNpAG69BV98PzV4HxlieMXYPgJJ/X916LwBtvQ5tux+kIFhb9gP4Zef16wFsc14/DOBaItKIaCeAS5zPDgF4DhHtICIN9o2xDV0KEe2A3bb0fgAbGWNTgP1gwNaEAPuBOCpsdsxZBgB/AeCjAAprcbyrQYvXoGfuh4jXoR5dfy8ALV+Htt0PUhCsLW8D8E4iehC2Sqg7yz8D+2HfC+DvAPwIgMEYmwfwDgD/DuCHAI4AMNb2kNuDM4P7KoD3MMaW6q0asIwR0W4AZzLGvrYax7cWtHoNeuV+aOI6hG2/G11+LwCtX4d23g8tNa+XNAdj7HEArwAAIjobwC86yw0Av8/XI6IfwSmqxxj7Bpx0cSK6AYC5tkfdOkQUg33Df4Ex9p/O4pNENMkYmyKiSQDTzvJj8M5qtgI4AeAyAJcQ0RHY9+0GIrqbMXbFWpxDq7TpGnT9/dDkdQijq+8FoG3XoW33g9QI1hAi2uD8VwB8EMAnnPdpIso4r6+CrQ086ttmBMDvAvjUOhz6iiEiAvBpAI8xxj4mfHQbgOud19cDuFVY/kYiSjhmsrMAPMAY+zhjbDNjbAdsx9kT3fLgt+saOPvq2vthBdchkG6+F4D2XQdnX+25H9bbg96rfwC+CGAKQAX2DO/tAN4NO0LgCQA3oprZvQPAz2E7je6EXSpW3M+jzt8b1/u8VnAdXgSAAdgH4CHn7xrYEQ93wdZ87gIwKmzzAdiRMj8H8KqAfe5AF0WKtPMadPP9sMLrcAR20EXOeY52dfO90O7r0K77QZaYkEgkkj5HmoYkEomkz5GCQCKRSPocKQgkEomkz5GCQCKRSPocKQgkEomkz5GCQCKpAxGZRPSQU93xYSL6AycPRFznViK6z3n9Smf9h5xqkT93Xn+OiK4gokXh84eI6Mr1OTOJpIoMH5VI6kBEOcZY1nm9AXa1y3sZY3/mLBsG8Ajs+O5rmF0kjm97N4A/Yoztdd5f4bx/9RqegkTSEKkRSCQRYXZl2BsAvMvJDgXsao/fAPAlAG9cr2OTSFpBCgKJpAkYY4dhPze8MuSbYGd3ftF53YgX+0xDZ6zSoUokkZFF5ySS5iEAIKKNAM4EcA9jjBGRQUTnM8bqdcz6oTQNSToNqRFIJE1ARKfDrvA4DeANAEYAPOVUwtwBaR6SdCFSEEgkESGiCdgVY/+R2VEWbwJwNWNsB7MrYV4CKQgkXYg0DUkk9UkR0UMAYrCbfnwewMeczlKnAfgxX5Ex9hQRLRHRCxhj94fs78XO/jh/yRj7yqocuUQSERk+KpFIJH2ONA1JJBJJnyMFgUQikfQ5UhBIJBJJnyMFgUQikfQ5UhBIJBJJnyMFgUQikfQ5UhBIJBJJn/P/AHIH0nQCsBhuAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "(comb['Dew'] - comb['Paciorek']).plot(title='Difference');"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Average monthly pattern from x13as adjustment"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 126,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-09-18T03:05:23.739234Z",
     "start_time": "2020-09-18T03:05:23.614542Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "((sm.observed - sm.seasadj).reset_index()\n",
    "                           .assign(MONTH = lambda x: x.DATE.dt.month)\n",
    "                           .groupby('MONTH')[0].mean().plot());"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Looking at individual months to figure out difference\n",
    "\n",
    "### Update: Try taking average of individual \"headship rate\"\n",
    "\n",
    "Assign an average headship to each person in the CPS, attempting to replicate Paciorek (2013, 2016)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 127,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-09-18T03:07:17.962314Z",
     "start_time": "2020-09-18T03:07:17.431135Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "May 2020\n",
      " Result (average): 49.42 \n",
      "\n",
      " Result (aggregate):  49.13 \n",
      "\n",
      " Goal:  49.12\n",
      "\n",
      " Households with incorrect headship sum: 0\n"
     ]
    }
   ],
   "source": [
    "date = '2020-05-01'\n",
    "cpsdt = pd.to_datetime(date).strftime('%b%y').lower()\n",
    "textdt = pd.to_datetime(date).strftime('%B %Y')\n",
    "print(textdt)\n",
    "\n",
    "# manually list out the IDs for series of interest \n",
    "var_names = ['PWSSWGT', 'QSTNUM', 'PRTAGE', 'PULINENO', \n",
    "             'PESPOUSE', 'HURESPLI'] \n",
    "\n",
    "dd = '2020_Basic_CPS_Public_Use_Record_Layout_plus_IO_Code_list.txt'\n",
    "data_dict = open(dd, 'r', encoding='iso-8859-1').read()\n",
    "\n",
    "p = f'\\n({\"|\".join(var_names)})\\s+(\\d+)\\s+.*?\\t+.*?(\\d\\d*).*?(\\d\\d+)'\n",
    "\n",
    "d = {s[0]: [int(s[2])-1, int(s[3]), f'{s[1]}s']\n",
    "     for s in re.findall(p, data_dict)}\n",
    "\n",
    "start, end, width = zip(*d.values())\n",
    "skip = ([f'{s - e}x' for s, e in zip(start, [0] + list(end[:-1]))])\n",
    "unpack_fmt = ''.join([j for i in zip(skip, width) for j in i])\n",
    "unpacker = struct.Struct(unpack_fmt).unpack_from    \n",
    "\n",
    "file = f'{cpsdt}pub.dat'\n",
    "raw_data = open(file, 'rb').readlines()\n",
    "data = [[*map(int, unpacker(row))] for row in raw_data]\n",
    "df = pd.DataFrame(data, columns=d.keys())\n",
    "\n",
    "hh16 = lambda x: np.where(x.PRTAGE > 15, 1, 0)\n",
    "\n",
    "hhsize = lambda x: x.groupby('QSTNUM').HH16.transform('sum') # count of age 16+\n",
    "\n",
    "head1 = (lambda x: (np.where((x.HURESPLI == x.PULINENO) & (x.HHSIZE == 1), 1, # alone (1)\n",
    "                   np.where((x.HURESPLI == x.PULINENO) & (x.PESPOUSE > 0), 0.5, # spouse1 (0.5)\n",
    "                   np.where((x.HURESPLI != -1) & (x.HURESPLI == x.PESPOUSE), 0.5, # spouse2 (0.5)\n",
    "                   np.nan)))))\n",
    "\n",
    "hhsum = lambda x: x.groupby('QSTNUM').HEAD1.transform('sum')\n",
    "\n",
    "head2 = (lambda x: (np.where(x.HEAD1.notnull(), np.nan, # already identified\n",
    "                    np.where(x.HH16 == 0, np.nan, # under 16\n",
    "                    np.where((x.HHSUM == 1) & (x.HEAD1.isnull()), 0, # living with family (0)\n",
    "                    1/x.HHSIZE))))) # roommates (1/nr)\n",
    "\n",
    "head = (lambda x: (np.where(x.HEAD1.notnull(), x.HEAD1, \n",
    "                   np.where(x.HEAD2.notnull(), x.HEAD2, np.nan))))\n",
    "\n",
    "hhsum2 = lambda x: x.groupby('QSTNUM').HEAD.transform('sum')\n",
    "\n",
    "droplist = ['HH16', 'HHSIZE', 'HEAD1', 'HHSUM', 'HEAD2', 'HURESPLI']\n",
    "\n",
    "data = df.assign(HH16 = hh16, \n",
    "                 HHSIZE = hhsize, \n",
    "                 HEAD1 = head1, \n",
    "                 HHSUM = hhsum, \n",
    "                 HEAD2 = head2, \n",
    "                 HEAD = head,\n",
    "                 HHSUM2 = hhsum2\n",
    "                ).drop(droplist, axis=1)\n",
    "\n",
    "d1 = data.query('PRTAGE > 15')\n",
    "result = np.average(d1.HEAD, weights=d1.PWSSWGT)\n",
    "print(f' Result (average): {result*100:.2f}',\n",
    "      '\\n\\n',f'Result (aggregate): ', comb.loc[date, 'Dew'].round(2),\n",
    "      '\\n\\n','Goal: ', comb.loc[date, 'Paciorek'])\n",
    "\n",
    "incorrect = len(d1[~d1.HHSUM2.between(0.99, 1.01)])\n",
    "print(f'\\n Households with incorrect headship sum: {incorrect}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-09-18T02:24:43.798969Z",
     "start_time": "2020-09-18T02:24:43.776072Z"
    }
   },
   "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>PRTAGE</th>\n",
       "      <th>PESPOUSE</th>\n",
       "      <th>PULINENO</th>\n",
       "      <th>PWSSWGT</th>\n",
       "      <th>QSTNUM</th>\n",
       "      <th>HEAD</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>72</td>\n",
       "      <td>-1</td>\n",
       "      <td>1</td>\n",
       "      <td>17479947</td>\n",
       "      <td>11906</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>80</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>17816015</td>\n",
       "      <td>11437</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>80</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>18182721</td>\n",
       "      <td>11437</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>50</td>\n",
       "      <td>-1</td>\n",
       "      <td>1</td>\n",
       "      <td>24249454</td>\n",
       "      <td>24833</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>85</td>\n",
       "      <td>-1</td>\n",
       "      <td>2</td>\n",
       "      <td>23134010</td>\n",
       "      <td>24833</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>123350</th>\n",
       "      <td>60</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>3273148</td>\n",
       "      <td>21248</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>123351</th>\n",
       "      <td>76</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>2613962</td>\n",
       "      <td>29576</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>123352</th>\n",
       "      <td>77</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>2532948</td>\n",
       "      <td>29576</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>123357</th>\n",
       "      <td>37</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>5620033</td>\n",
       "      <td>36237</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>123358</th>\n",
       "      <td>35</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>8175390</td>\n",
       "      <td>36237</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>76441 rows × 6 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        PRTAGE  PESPOUSE  PULINENO   PWSSWGT  QSTNUM  HEAD\n",
       "0           72        -1         1  17479947   11906   1.0\n",
       "1           80         2         1  17816015   11437   0.5\n",
       "2           80         1         2  18182721   11437   0.5\n",
       "4           50        -1         1  24249454   24833   0.5\n",
       "5           85        -1         2  23134010   24833   0.5\n",
       "...        ...       ...       ...       ...     ...   ...\n",
       "123350      60         1         2   3273148   21248   0.5\n",
       "123351      76         2         1   2613962   29576   0.5\n",
       "123352      77         1         2   2532948   29576   0.5\n",
       "123357      37         2         1   5620033   36237   0.5\n",
       "123358      35         1         2   8175390   36237   0.5\n",
       "\n",
       "[76441 rows x 6 columns]"
      ]
     },
     "execution_count": 79,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "d1"
   ]
  },
  {
   "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": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-09-11T23:37:04.244044Z",
     "start_time": "2020-09-11T23:37:04.237842Z"
    }
   },
   "source": [
    "### Living with Family -- Does that match?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-09-15T05:18:07.636690Z",
     "start_time": "2020-09-15T05:18:07.627495Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "36.39200602822098"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(data.query('HWHHWTLN != PULINENO and PERRP in [48, 49, 50, 51, 52, 53]').PWSSWGT.sum() / \n",
    " data.PWSSWGT.sum()) * 100"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Goal: 20.76"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-09-15T05:18:07.646736Z",
     "start_time": "2020-09-15T05:18:07.637634Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "18.20508384690031"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(data.query('PRFAMREL not in [0, 1, 2]').PWCMPWGT.sum() / \n",
    " data.PWCMPWGT.sum()) * 100"
   ]
  },
  {
   "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": []
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-09-15T16:51:36.829131Z",
     "start_time": "2020-09-15T16:51:36.827183Z"
    }
   },
   "outputs": [],
   "source": [
    "# Working area\n",
    "\n",
    "#d1['HHEAD'] = d1.groupby('QSTNUM').HEAD.transform('sum')\n",
    "#d1[d1.HHEAD < 0.99]\n",
    "#d1.query('QSTNUM == 37931')\n",
    "\n",
    "\n",
    "    # More detailed headship indicator for those 16+\n",
    "    #spcheck = lambda x: 1 if x.HURESPLI.iloc[0] in x.SPOUSE.to_list() else 0\n",
    "    #dfm['SPCHECK'] = dfm.QSTNUM.map(dfm.groupby('QSTNUM').apply(spcheck).to_dict())\n",
    "    \n",
    "    # identify husband-wife household\n",
    "    #hhcat = lambda x: np.where((x.HRHTYPE.isin([1, 2])) & (x.SPCHECK == 1), 1, 0)\n",
    "\n",
    "    # identify those age 16 or older\n",
    "    #hh16 = lambda x: np.where(x.AGE > 15, 1, 0)\n",
    "\n",
    "    # count of age 16+\n",
    "    #hhsize = lambda x: x.groupby('QSTNUM').HH16.transform('sum') \n",
    "\n",
    "    # from Paciorek (2013, 2016)\n",
    "    #head = (lambda x: (np.where((x.HURESPLI == x.PULINENO) & (x.HHSIZE == 1), 1, \n",
    "    #                   np.where((x.HURESPLI == x.PULINENO) & (x.SPOUSE > 0), 0.5, \n",
    "    #                   np.where(x.HURESPLI == x.SPOUSE, 0.5, \n",
    "    #                   np.where((x.HHCAT == 1) & (x.HH16 == 1), 0,  \n",
    "    #                   np.where(x.HH16 == 1, 1/x.HHSIZE, np.nan))))))) \n",
    "\n",
    "    #dfm = (dfm.assign(HHCAT=hhcat, HH16=hh16, HHSIZE=hhsize, HEAD=head)\n",
    "    #          .drop(['HHCAT', 'HH16', 'HHSIZE', 'HURESPLI', 'HRHTYPE'], axis=1))\n",
    "    \n",
    "    \n",
    "    \n",
    "#sp1 = df[df.PESPOUSE > 0].groupby('QSTNUM').PESPOUSE.max()\n",
    "#sp1.name = 'SPOUSE1'\n",
    "#sp2 = df[df.PESPOUSE > 0].groupby('QSTNUM').PESPOUSE.min()\n",
    "#sp2.name = 'SPOUSE2'\n",
    "#df = df.merge(sp1.reset_index()).merge(sp2.reset_index())"
   ]
  }
 ],
 "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": "3.8.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}