GaussianMixture_RealEstatePrices

GaussianMixture.ipynb

gistfile1.txt

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import matplotlib.dates as mdates\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "from sklearn.linear_model import LinearRegression\n",
    "import scipy\n",
    "from scipy.stats import gamma\n",
    "from scipy.stats import norm\n",
    "from scipy.stats import lognorm\n",
    "import pickle"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "DOWNLOAD DATA FROM\n",
    "\n",
    "https://www.kaggle.com/hm-land-registry/uk-housing-prices-paid\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Selected columns\n",
    "columns = [\n",
    "       'Price', 'Date of Transfer',\n",
    "       'Property Type', 'Old/New', 'Duration', 'Town/City', 'District',\n",
    "       'County', 'PPDCategory Type'\n",
    "       ]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Min date: 2012-01-01 00:00\n",
      "Min max: 2014-12-30 00:00\n"
     ]
    }
   ],
   "source": [
    "# Read part of the dataset\n",
    "df = pd.read_csv(\"data/train_price_houses.csv\", usecols=columns)\n",
    "print(\"Min date: {}\".format(df['Date of Transfer'].min()))\n",
    "print(\"Min max: {}\".format(df['Date of Transfer'].max()))\n",
    "df['Date of Transfer'] = pd.DatetimeIndex(df['Date of Transfer'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(2461243, 9)"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Adjust price for inflation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "df['count_col'] = 0\n",
    "df['Price_mean'] = df['Price'] \n",
    "df['Price_std'] = df['Price'] \n",
    "df_group = df.groupby(by=['Town/City', 'County']).agg({'count_col': 'count', \n",
    "                                                       'Price_mean' : 'mean', 'Price_std': 'std'})\n",
    "df_group = df_group.dropna()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_group_date = df.groupby(by='Date of Transfer').agg({'Price_mean': 'mean', 'count_col': 'count'}).reset_index()\n",
    "df_group_date['Date of Transfer'] = pd.to_datetime(df_group_date['Date of Transfer'])\n",
    "\n",
    "# One week rolling window\n",
    "df_group_date['Price_x_count'] = df_group_date['Price_mean']*df_group_date['count_col']\n",
    "df_group_date['Price_sum'] = df_group_date['Price_x_count'].rolling(window=20, min_periods=7).sum()\n",
    "df_group_date['count_sum'] = df_group_date['count_col'].rolling(window=20, min_periods=7).sum()\n",
    "df_group_date['Price_mean'] = df_group_date['Price_sum']/df_group_date['count_sum']\n",
    "\n",
    "df_group_date = df_group_date.dropna().reset_index()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "x_train = np.array(df_group_date.index, dtype=np.float32)\n",
    "y_train = np.array(df_group_date['Price_mean'])\n",
    "\n",
    "params, _ = scipy.optimize.curve_fit(lambda t,a,b: a*np.exp(b*t),  x_train,  y_train, p0=(2e5, 0.01))\n",
    "alpha, beta = params[0], params[1]\n",
    "y_pred = alpha*np.exp(beta*x_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.00022242820326361787"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "beta"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "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": [
    "locator = mdates.YearLocator(1)\n",
    "locator_fmt = mdates.DateFormatter(\"%Y\")\n",
    "fig, ax = plt.subplots()\n",
    "ax.plot(df_group_date['Date of Transfer'], df_group_date['Price_mean'])\n",
    "ax.xaxis.set_major_locator(locator)\n",
    "ax.xaxis.set_major_formatter(locator_fmt)\n",
    "plt.yticks([220e3, 240e3, 260e3, 280e3, 300e3], ['220k', '240k', '260k', '280k', '300k'])\n",
    "plt.xlabel('Year')\n",
    "plt.ylabel('Mean price (£)')\n",
    "plt.grid()\n",
    "plt.tight_layout()\n",
    "plt.savefig('figures_mixture/prices.png', dpi=600)"
   ]
  },
  {
   "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": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "locator = mdates.YearLocator(1)\n",
    "locator_fmt = mdates.DateFormatter(\"%Y\")\n",
    "fig, ax = plt.subplots()\n",
    "ax.plot(df_group_date['Date of Transfer'], df_group_date['Price_mean'])\n",
    "ax.plot(df_group_date['Date of Transfer'], y_pred, color='r')\n",
    "ax.xaxis.set_major_locator(locator)\n",
    "ax.xaxis.set_major_formatter(locator_fmt)\n",
    "plt.yticks([220e3, 240e3, 260e3, 280e3, 300e3], ['220k', '240k', '260k', '280k', '300k'])\n",
    "plt.xlabel('Year')\n",
    "plt.ylabel('Mean price (£)')\n",
    "plt.grid()\n",
    "plt.legend(['Mean prices', 'Exponential fit'])\n",
    "plt.tight_layout()\n",
    "plt.savefig('figures_mixture/prices_fit.png', dpi=600)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Correction factor\n",
    "ini_date = df_group_date['Date of Transfer'][0]\n",
    "df_group_date['factor'] = np.exp(-beta*(df_group_date['Date of Transfer'] - ini_date).dt.days)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "try:\n",
    "    df = df.drop(columns=['factor'])\n",
    "except:\n",
    "    pass\n",
    "df = pd.merge(df, df_group_date[['Date of Transfer', 'factor']], on='Date of Transfer', how='left')\n",
    "df['Price_adj'] = df['Price']*df['factor']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Plot Adjusted Price"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_group_adj = df[df['Date of Transfer'] > ini_date].groupby(by='Date of Transfer').agg({'Price_adj': 'mean', 'count_col': 'count'}).reset_index()\n",
    "df_group_adj['Date of Transfer'] = pd.to_datetime(df_group_adj['Date of Transfer'])\n",
    "\n",
    "# One week rolling window\n",
    "df_group_adj['Price_x_count'] = df_group_adj['Price_adj']*df_group_adj['count_col']\n",
    "df_group_adj['Price_sum'] = df_group_adj['Price_x_count'].rolling(window=20, min_periods=7).sum()\n",
    "df_group_adj['count_sum'] = df_group_adj['count_col'].rolling(window=20, min_periods=7).sum()\n",
    "df_group_adj['Price_adj'] = df_group_adj['Price_sum']/df_group_adj['count_sum']\n",
    "df_group_adj = df_group_adj.dropna().reset_index()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "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": [
    "locator = mdates.YearLocator(1)\n",
    "locator_fmt = mdates.DateFormatter(\"%Y\")\n",
    "fig, ax = plt.subplots()\n",
    "ax.plot(df_group_adj['Date of Transfer'], df_group_adj['Price_adj'])\n",
    "plt.hlines(alpha, df_group_adj['Date of Transfer'].min(), df_group_adj['Date of Transfer'].max(), color='r')\n",
    "ax.xaxis.set_major_locator(locator)\n",
    "ax.xaxis.set_major_formatter(locator_fmt)\n",
    "plt.yticks([210e3, 220e3, 230e3, 240e3, 250e3, 260e3], ['210k', '220k', '230k', '240k', '250k', '260k'])\n",
    "plt.xlabel('Year')\n",
    "plt.ylabel('Mean price (£)')\n",
    "plt.grid()\n",
    "plt.legend(['Adjusted prices', 'Mean price'])\n",
    "plt.tight_layout()\n",
    "plt.savefig('figures_mixture/prices_adjusted.png', dpi=600)\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Price Distribution"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "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": [
    "# Split between old and new\n",
    "col = 'County'\n",
    "set1 = ['GREATER LONDON']\n",
    "set2 = ['WEST MIDLANDS', 'GREATER MANCHESTER']\n",
    "df_set1 = df[df[col].isin(set1)]\n",
    "df_set2 = df[df[col].isin(set2)]\n",
    "df_subset = df[df[col].isin(set1+set2)]\n",
    "#df_old = df[df['Property Type'] == 'D']\n",
    "#df_new = df[df['Property Type'] != 'D']\n",
    "#PPDCategory Type\n",
    "\n",
    "plt.figure()\n",
    "plt.hist(df_subset['Price_adj'], bins=np.arange(0, 1.2e6, 1e3), alpha=.5)\n",
    "plt.hist(df_set1['Price_adj'], bins=np.arange(0, 1.2e6, 1e3), alpha=.5)\n",
    "plt.hist(df_set2['Price_adj'], bins=np.arange(0, 1.2e6, 1e3), alpha=.5)\n",
    "#plt.hist(np.log(df['Price_adj']), bins=np.arange(8, 16, 0.01), alpha=.5)\n",
    "#plt.hist(np.log(df_old['Price_adj']), bins=np.arange(8, 16, 0.01), alpha=.5)\n",
    "#plt.hist(np.log(df_new['Price_adj']), bins=np.arange(8, 16, 0.01), alpha=.5)\n",
    "#plt.yscale('log')\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "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": [
    "plt.figure()\n",
    "plt.hist(df['Price_adj'], bins=np.arange(0, 1.2e6, 1e3), alpha=1)\n",
    "plt.xticks([0, 2e5, 4e5, 6e5, 8e5, 1e6, 1.2e6], ['0', '200k', '400k', '600k', '800k', '1M', '1.2M'])\n",
    "plt.yticks([0, 3e3, 6e3, 9e3, 12e3], ['0', '3k', '6k', '9k', '12k'])\n",
    "plt.xlabel('Price (£)')\n",
    "plt.ylabel('Frequency')\n",
    "plt.grid()\n",
    "plt.tight_layout()\n",
    "plt.savefig('figures_mixture/frequencies_price.png', dpi=600)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# plt.figure()\n",
    "# plt.hist(df['Price_adj'], bins=np.arange(0, 1e8, 5e5), alpha=1, log=True)\n",
    "# plt.xticks([0, 2e7, 4e7, 6e7, 8e7, 10e7], ['0', '20M', '40M', '60M', '80M', '100M'])\n",
    "# plt.yticks([1, 1e2, 1e4, 1e6], ['1', '100', '10k', '1M'])\n",
    "# plt.grid()\n",
    "# plt.xlabel('Price (£)')\n",
    "# plt.ylabel('Frequency')\n",
    "# plt.show()\n",
    "\n",
    "bins = np.arange(0, 1.2e6, 1e3)\n",
    "x = df['Price_adj']\n",
    "log_x = np.log(x)\n",
    "log_bins = np.arange(8, 19, 0.01)\n",
    "\n",
    "fig, (ax1, ax2) = plt.subplots(1, 2)\n",
    "fig.set_figwidth(10)\n",
    "\n",
    "# Plot 1\n",
    "ax1.hist(log_x, bins=log_bins, alpha=1, log=True)\n",
    "plt.setp(ax1, xticks=np.log(10**np.array([4, 5, 6, 7, 8])), xticklabels=['10k', '100k', '1M', '10M', '100M'])\n",
    "#plt.yticks([0, 4e3, 8e3, 12e3, 16e3], ['0', '4k', '8k', '12k', '16k'])\n",
    "ax1.grid()\n",
    "plt.setp(ax1, xlabel='Price (£)')\n",
    "plt.setp(ax1, ylabel='Frequency')\n",
    "\n",
    "# Plot 2\n",
    "ax2.hist(log_x, bins=log_bins, alpha=1)\n",
    "plt.setp(ax2, xticks=np.log(10**np.array([4, 5, 6, 7, 8])), xticklabels=['10k', '100k', '1M', '10M', '100M'])\n",
    "plt.setp(ax2, yticks=[0, 4e3, 8e3, 12e3, 16e3], yticklabels=['0', '4k', '8k', '12k', '16k'])\n",
    "ax2.grid()\n",
    "plt.setp(ax2, xlabel='Price (£)')\n",
    "plt.setp(ax2, ylabel='Frequency')\n",
    "\n",
    "plt.savefig('figures_mixture/frequencies_log_price.png', dpi=600)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "290114.4940007791\n",
      "114095.06862380102\n",
      "1142955.227998204\n"
     ]
    }
   ],
   "source": [
    "print(df_set1['Price_adj'].median())\n",
    "print(df_set2['Price_adj'].median())\n",
    "print(df['Price_adj'].quantile(.99))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Data Exploration"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['S' 'D' 'T' 'F' 'O']\n",
      "['N' 'Y']\n",
      "['F' 'L']\n",
      "1150\n",
      "349\n",
      "113\n",
      "['A' 'B']\n"
     ]
    }
   ],
   "source": [
    "print(df['Property Type'].unique())\n",
    "print(df['Old/New'].unique())\n",
    "print(df['Duration'].unique())\n",
    "print(len(df['Town/City'].unique()))\n",
    "print(len(df['District'].unique()))\n",
    "print(len(df['County'].unique()))\n",
    "print(df['PPDCategory Type'].unique())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "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>County</th>\n",
       "      <th>Price_adj</th>\n",
       "      <th>count_col</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>BLAENAU GWENT</td>\n",
       "      <td>73269.808797</td>\n",
       "      <td>2016</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>MERTHYR TYDFIL</td>\n",
       "      <td>92139.130742</td>\n",
       "      <td>1817</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>CITY OF KINGSTON UPON HULL</td>\n",
       "      <td>93040.445431</td>\n",
       "      <td>8930</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>STOKE-ON-TRENT</td>\n",
       "      <td>93357.951855</td>\n",
       "      <td>8884</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>CHESHIRE</td>\n",
       "      <td>93582.763717</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>RHONDDA CYNON TAFF</td>\n",
       "      <td>97942.358842</td>\n",
       "      <td>8668</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>NEATH PORT TALBOT</td>\n",
       "      <td>98825.860753</td>\n",
       "      <td>4795</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>BLACKPOOL</td>\n",
       "      <td>102804.562947</td>\n",
       "      <td>5346</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>BLACKBURN WITH DARWEN</td>\n",
       "      <td>104437.944282</td>\n",
       "      <td>4375</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>HARTLEPOOL</td>\n",
       "      <td>106497.267459</td>\n",
       "      <td>3278</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>NORTH EAST LINCOLNSHIRE</td>\n",
       "      <td>107329.134530</td>\n",
       "      <td>6363</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>COUNTY DURHAM</td>\n",
       "      <td>109642.945978</td>\n",
       "      <td>19041</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>CAERPHILLY</td>\n",
       "      <td>111230.417326</td>\n",
       "      <td>6052</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>REDCAR AND CLEVELAND</td>\n",
       "      <td>113482.892914</td>\n",
       "      <td>5042</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>MIDDLESBROUGH</td>\n",
       "      <td>116489.333010</td>\n",
       "      <td>4511</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>NORTH LINCOLNSHIRE</td>\n",
       "      <td>117873.960134</td>\n",
       "      <td>6222</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>TORFAEN</td>\n",
       "      <td>119174.545248</td>\n",
       "      <td>2935</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>DARLINGTON</td>\n",
       "      <td>122229.516852</td>\n",
       "      <td>4389</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>CARMARTHENSHIRE</td>\n",
       "      <td>125318.178312</td>\n",
       "      <td>6601</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>BRIDGEND</td>\n",
       "      <td>126621.160547</td>\n",
       "      <td>5795</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>HALTON</td>\n",
       "      <td>127586.846821</td>\n",
       "      <td>4083</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>MERSEYSIDE</td>\n",
       "      <td>127900.100493</td>\n",
       "      <td>46197</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>DENBIGHSHIRE</td>\n",
       "      <td>128796.602233</td>\n",
       "      <td>3645</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>SOUTH YORKSHIRE</td>\n",
       "      <td>132634.798086</td>\n",
       "      <td>48310</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>TYNE AND WEAR</td>\n",
       "      <td>132836.084261</td>\n",
       "      <td>39882</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>STOCKTON-ON-TEES</td>\n",
       "      <td>133128.442604</td>\n",
       "      <td>7856</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>LANCASHIRE</td>\n",
       "      <td>136263.712622</td>\n",
       "      <td>48494</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27</th>\n",
       "      <td>LEICESTER</td>\n",
       "      <td>136328.622265</td>\n",
       "      <td>8937</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28</th>\n",
       "      <td>CITY OF NOTTINGHAM</td>\n",
       "      <td>136861.714621</td>\n",
       "      <td>9726</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29</th>\n",
       "      <td>WREXHAM</td>\n",
       "      <td>137288.756408</td>\n",
       "      <td>4211</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30</th>\n",
       "      <td>WREKIN</td>\n",
       "      <td>138486.625136</td>\n",
       "      <td>6979</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>31</th>\n",
       "      <td>NEWPORT</td>\n",
       "      <td>139437.992324</td>\n",
       "      <td>5377</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>32</th>\n",
       "      <td>SWANSEA</td>\n",
       "      <td>139771.051919</td>\n",
       "      <td>8619</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>33</th>\n",
       "      <td>WEST YORKSHIRE</td>\n",
       "      <td>140539.665257</td>\n",
       "      <td>85211</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>34</th>\n",
       "      <td>CITY OF DERBY</td>\n",
       "      <td>141649.405825</td>\n",
       "      <td>10028</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35</th>\n",
       "      <td>ISLE OF ANGLESEY</td>\n",
       "      <td>143291.146056</td>\n",
       "      <td>2690</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>36</th>\n",
       "      <td>LINCOLNSHIRE</td>\n",
       "      <td>143491.347628</td>\n",
       "      <td>36561</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>37</th>\n",
       "      <td>NOTTINGHAMSHIRE</td>\n",
       "      <td>143506.978333</td>\n",
       "      <td>36573</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>38</th>\n",
       "      <td>CITY OF PLYMOUTH</td>\n",
       "      <td>144657.571347</td>\n",
       "      <td>11370</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>39</th>\n",
       "      <td>GREATER MANCHESTER</td>\n",
       "      <td>145231.369334</td>\n",
       "      <td>97432</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>40</th>\n",
       "      <td>CONWY</td>\n",
       "      <td>145332.987587</td>\n",
       "      <td>5180</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>41</th>\n",
       "      <td>GWYNEDD</td>\n",
       "      <td>147036.995239</td>\n",
       "      <td>4208</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>42</th>\n",
       "      <td>DERBYSHIRE</td>\n",
       "      <td>147294.649761</td>\n",
       "      <td>33630</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>43</th>\n",
       "      <td>EAST RIDING OF YORKSHIRE</td>\n",
       "      <td>148328.992704</td>\n",
       "      <td>16176</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>44</th>\n",
       "      <td>WEST MIDLANDS</td>\n",
       "      <td>149291.498659</td>\n",
       "      <td>88194</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45</th>\n",
       "      <td>LUTON</td>\n",
       "      <td>149538.754231</td>\n",
       "      <td>6908</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>46</th>\n",
       "      <td>CUMBRIA</td>\n",
       "      <td>151739.284148</td>\n",
       "      <td>21670</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>47</th>\n",
       "      <td>NORTHUMBERLAND</td>\n",
       "      <td>151911.398850</td>\n",
       "      <td>12656</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>48</th>\n",
       "      <td>CITY OF PETERBOROUGH</td>\n",
       "      <td>152278.652859</td>\n",
       "      <td>8150</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49</th>\n",
       "      <td>PEMBROKESHIRE</td>\n",
       "      <td>152330.039110</td>\n",
       "      <td>4493</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50</th>\n",
       "      <td>POWYS</td>\n",
       "      <td>154791.276240</td>\n",
       "      <td>4342</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>51</th>\n",
       "      <td>FLINTSHIRE</td>\n",
       "      <td>155422.313028</td>\n",
       "      <td>5682</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>52</th>\n",
       "      <td>PORTSMOUTH</td>\n",
       "      <td>156280.267976</td>\n",
       "      <td>9020</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53</th>\n",
       "      <td>SWINDON</td>\n",
       "      <td>157336.880129</td>\n",
       "      <td>10587</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>54</th>\n",
       "      <td>STAFFORDSHIRE</td>\n",
       "      <td>159243.168849</td>\n",
       "      <td>34041</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>55</th>\n",
       "      <td>TORBAY</td>\n",
       "      <td>162658.155764</td>\n",
       "      <td>7395</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>56</th>\n",
       "      <td>SOUTHAMPTON</td>\n",
       "      <td>164225.810983</td>\n",
       "      <td>9877</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>57</th>\n",
       "      <td>MEDWAY</td>\n",
       "      <td>164857.459768</td>\n",
       "      <td>11829</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>58</th>\n",
       "      <td>WARRINGTON</td>\n",
       "      <td>165903.758442</td>\n",
       "      <td>8687</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>59</th>\n",
       "      <td>CEREDIGION</td>\n",
       "      <td>165913.966997</td>\n",
       "      <td>2371</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>60</th>\n",
       "      <td>NORTHAMPTONSHIRE</td>\n",
       "      <td>167826.572749</td>\n",
       "      <td>35029</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>61</th>\n",
       "      <td>THURROCK</td>\n",
       "      <td>169655.694670</td>\n",
       "      <td>6808</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>62</th>\n",
       "      <td>NORFOLK</td>\n",
       "      <td>171960.048301</td>\n",
       "      <td>45387</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>63</th>\n",
       "      <td>CARDIFF</td>\n",
       "      <td>175319.455507</td>\n",
       "      <td>14751</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>64</th>\n",
       "      <td>ISLE OF WIGHT</td>\n",
       "      <td>177023.087517</td>\n",
       "      <td>8017</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>65</th>\n",
       "      <td>LEICESTERSHIRE</td>\n",
       "      <td>178787.594166</td>\n",
       "      <td>32661</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>66</th>\n",
       "      <td>SHROPSHIRE</td>\n",
       "      <td>180224.994285</td>\n",
       "      <td>12745</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>67</th>\n",
       "      <td>SUFFOLK</td>\n",
       "      <td>186618.696740</td>\n",
       "      <td>37160</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>68</th>\n",
       "      <td>THE VALE OF GLAMORGAN</td>\n",
       "      <td>186984.017697</td>\n",
       "      <td>5470</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>69</th>\n",
       "      <td>CHESHIRE WEST AND CHESTER</td>\n",
       "      <td>187133.783921</td>\n",
       "      <td>14340</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>70</th>\n",
       "      <td>SOMERSET</td>\n",
       "      <td>190042.943176</td>\n",
       "      <td>27498</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>71</th>\n",
       "      <td>WORCESTERSHIRE</td>\n",
       "      <td>190473.349404</td>\n",
       "      <td>26388</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>72</th>\n",
       "      <td>HEREFORDSHIRE</td>\n",
       "      <td>191937.252028</td>\n",
       "      <td>7669</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>73</th>\n",
       "      <td>NORTH YORKSHIRE</td>\n",
       "      <td>192596.686729</td>\n",
       "      <td>28848</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>74</th>\n",
       "      <td>SOUTH GLOUCESTERSHIRE</td>\n",
       "      <td>194889.125019</td>\n",
       "      <td>12794</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75</th>\n",
       "      <td>CORNWALL</td>\n",
       "      <td>199168.369268</td>\n",
       "      <td>27400</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>76</th>\n",
       "      <td>BEDFORD</td>\n",
       "      <td>199342.258983</td>\n",
       "      <td>7971</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>77</th>\n",
       "      <td>YORK</td>\n",
       "      <td>199457.010277</td>\n",
       "      <td>10186</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>78</th>\n",
       "      <td>MONMOUTHSHIRE</td>\n",
       "      <td>200824.646281</td>\n",
       "      <td>3956</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>79</th>\n",
       "      <td>NORTH SOMERSET</td>\n",
       "      <td>201204.299395</td>\n",
       "      <td>11683</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>80</th>\n",
       "      <td>BOURNEMOUTH</td>\n",
       "      <td>202581.476247</td>\n",
       "      <td>10514</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>81</th>\n",
       "      <td>SOUTHEND-ON-SEA</td>\n",
       "      <td>202829.761767</td>\n",
       "      <td>9059</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>82</th>\n",
       "      <td>CHESHIRE EAST</td>\n",
       "      <td>203600.412792</td>\n",
       "      <td>17859</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>83</th>\n",
       "      <td>CITY OF BRISTOL</td>\n",
       "      <td>204345.825712</td>\n",
       "      <td>21591</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>84</th>\n",
       "      <td>WARWICKSHIRE</td>\n",
       "      <td>205537.558500</td>\n",
       "      <td>25625</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>85</th>\n",
       "      <td>SLOUGH</td>\n",
       "      <td>207672.811222</td>\n",
       "      <td>4803</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>86</th>\n",
       "      <td>CENTRAL BEDFORDSHIRE</td>\n",
       "      <td>208143.014559</td>\n",
       "      <td>14946</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>87</th>\n",
       "      <td>MILTON KEYNES</td>\n",
       "      <td>210914.827990</td>\n",
       "      <td>12499</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>88</th>\n",
       "      <td>DEVON</td>\n",
       "      <td>212786.196210</td>\n",
       "      <td>41792</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>89</th>\n",
       "      <td>GLOUCESTERSHIRE</td>\n",
       "      <td>213038.633608</td>\n",
       "      <td>31422</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>90</th>\n",
       "      <td>WILTSHIRE</td>\n",
       "      <td>217053.701071</td>\n",
       "      <td>23671</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>91</th>\n",
       "      <td>KENT</td>\n",
       "      <td>222816.214665</td>\n",
       "      <td>73717</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>92</th>\n",
       "      <td>CAMBRIDGESHIRE</td>\n",
       "      <td>226680.603347</td>\n",
       "      <td>32160</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>93</th>\n",
       "      <td>EAST SUSSEX</td>\n",
       "      <td>229368.771959</td>\n",
       "      <td>31054</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>94</th>\n",
       "      <td>ESSEX</td>\n",
       "      <td>234025.117029</td>\n",
       "      <td>70874</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>95</th>\n",
       "      <td>RUTLAND</td>\n",
       "      <td>235240.240618</td>\n",
       "      <td>1957</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>96</th>\n",
       "      <td>DORSET</td>\n",
       "      <td>239313.188820</td>\n",
       "      <td>22762</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>97</th>\n",
       "      <td>READING</td>\n",
       "      <td>243440.568329</td>\n",
       "      <td>7985</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>98</th>\n",
       "      <td>HAMPSHIRE</td>\n",
       "      <td>250628.699991</td>\n",
       "      <td>67579</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>99</th>\n",
       "      <td>WEST SUSSEX</td>\n",
       "      <td>255318.828667</td>\n",
       "      <td>45916</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>100</th>\n",
       "      <td>BRACKNELL FOREST</td>\n",
       "      <td>264293.307341</td>\n",
       "      <td>6143</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>101</th>\n",
       "      <td>POOLE</td>\n",
       "      <td>265033.774489</td>\n",
       "      <td>8114</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>102</th>\n",
       "      <td>BRIGHTON AND HOVE</td>\n",
       "      <td>276522.260494</td>\n",
       "      <td>14962</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>103</th>\n",
       "      <td>BATH AND NORTH EAST SOMERSET</td>\n",
       "      <td>276606.562034</td>\n",
       "      <td>9021</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>104</th>\n",
       "      <td>WEST BERKSHIRE</td>\n",
       "      <td>281778.408823</td>\n",
       "      <td>7781</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>105</th>\n",
       "      <td>ISLES OF SCILLY</td>\n",
       "      <td>282605.156667</td>\n",
       "      <td>45</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>106</th>\n",
       "      <td>OXFORDSHIRE</td>\n",
       "      <td>295648.926991</td>\n",
       "      <td>30887</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>107</th>\n",
       "      <td>WOKINGHAM</td>\n",
       "      <td>306010.842833</td>\n",
       "      <td>8156</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>108</th>\n",
       "      <td>HERTFORDSHIRE</td>\n",
       "      <td>306911.661768</td>\n",
       "      <td>56536</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>109</th>\n",
       "      <td>BUCKINGHAMSHIRE</td>\n",
       "      <td>333595.258420</td>\n",
       "      <td>26067</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>110</th>\n",
       "      <td>SURREY</td>\n",
       "      <td>386774.070890</td>\n",
       "      <td>61605</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>111</th>\n",
       "      <td>WINDSOR AND MAIDENHEAD</td>\n",
       "      <td>442982.051869</td>\n",
       "      <td>7118</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>112</th>\n",
       "      <td>GREATER LONDON</td>\n",
       "      <td>452225.698396</td>\n",
       "      <td>337924</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                           County      Price_adj  count_col\n",
       "0                   BLAENAU GWENT   73269.808797       2016\n",
       "1                  MERTHYR TYDFIL   92139.130742       1817\n",
       "2      CITY OF KINGSTON UPON HULL   93040.445431       8930\n",
       "3                  STOKE-ON-TRENT   93357.951855       8884\n",
       "4                        CHESHIRE   93582.763717          1\n",
       "5              RHONDDA CYNON TAFF   97942.358842       8668\n",
       "6               NEATH PORT TALBOT   98825.860753       4795\n",
       "7                       BLACKPOOL  102804.562947       5346\n",
       "8           BLACKBURN WITH DARWEN  104437.944282       4375\n",
       "9                      HARTLEPOOL  106497.267459       3278\n",
       "10        NORTH EAST LINCOLNSHIRE  107329.134530       6363\n",
       "11                  COUNTY DURHAM  109642.945978      19041\n",
       "12                     CAERPHILLY  111230.417326       6052\n",
       "13           REDCAR AND CLEVELAND  113482.892914       5042\n",
       "14                  MIDDLESBROUGH  116489.333010       4511\n",
       "15             NORTH LINCOLNSHIRE  117873.960134       6222\n",
       "16                        TORFAEN  119174.545248       2935\n",
       "17                     DARLINGTON  122229.516852       4389\n",
       "18                CARMARTHENSHIRE  125318.178312       6601\n",
       "19                       BRIDGEND  126621.160547       5795\n",
       "20                         HALTON  127586.846821       4083\n",
       "21                     MERSEYSIDE  127900.100493      46197\n",
       "22                   DENBIGHSHIRE  128796.602233       3645\n",
       "23                SOUTH YORKSHIRE  132634.798086      48310\n",
       "24                  TYNE AND WEAR  132836.084261      39882\n",
       "25               STOCKTON-ON-TEES  133128.442604       7856\n",
       "26                     LANCASHIRE  136263.712622      48494\n",
       "27                      LEICESTER  136328.622265       8937\n",
       "28             CITY OF NOTTINGHAM  136861.714621       9726\n",
       "29                        WREXHAM  137288.756408       4211\n",
       "30                         WREKIN  138486.625136       6979\n",
       "31                        NEWPORT  139437.992324       5377\n",
       "32                        SWANSEA  139771.051919       8619\n",
       "33                 WEST YORKSHIRE  140539.665257      85211\n",
       "34                  CITY OF DERBY  141649.405825      10028\n",
       "35               ISLE OF ANGLESEY  143291.146056       2690\n",
       "36                   LINCOLNSHIRE  143491.347628      36561\n",
       "37                NOTTINGHAMSHIRE  143506.978333      36573\n",
       "38               CITY OF PLYMOUTH  144657.571347      11370\n",
       "39             GREATER MANCHESTER  145231.369334      97432\n",
       "40                          CONWY  145332.987587       5180\n",
       "41                        GWYNEDD  147036.995239       4208\n",
       "42                     DERBYSHIRE  147294.649761      33630\n",
       "43       EAST RIDING OF YORKSHIRE  148328.992704      16176\n",
       "44                  WEST MIDLANDS  149291.498659      88194\n",
       "45                          LUTON  149538.754231       6908\n",
       "46                        CUMBRIA  151739.284148      21670\n",
       "47                 NORTHUMBERLAND  151911.398850      12656\n",
       "48           CITY OF PETERBOROUGH  152278.652859       8150\n",
       "49                  PEMBROKESHIRE  152330.039110       4493\n",
       "50                          POWYS  154791.276240       4342\n",
       "51                     FLINTSHIRE  155422.313028       5682\n",
       "52                     PORTSMOUTH  156280.267976       9020\n",
       "53                        SWINDON  157336.880129      10587\n",
       "54                  STAFFORDSHIRE  159243.168849      34041\n",
       "55                         TORBAY  162658.155764       7395\n",
       "56                    SOUTHAMPTON  164225.810983       9877\n",
       "57                         MEDWAY  164857.459768      11829\n",
       "58                     WARRINGTON  165903.758442       8687\n",
       "59                     CEREDIGION  165913.966997       2371\n",
       "60               NORTHAMPTONSHIRE  167826.572749      35029\n",
       "61                       THURROCK  169655.694670       6808\n",
       "62                        NORFOLK  171960.048301      45387\n",
       "63                        CARDIFF  175319.455507      14751\n",
       "64                  ISLE OF WIGHT  177023.087517       8017\n",
       "65                 LEICESTERSHIRE  178787.594166      32661\n",
       "66                     SHROPSHIRE  180224.994285      12745\n",
       "67                        SUFFOLK  186618.696740      37160\n",
       "68          THE VALE OF GLAMORGAN  186984.017697       5470\n",
       "69      CHESHIRE WEST AND CHESTER  187133.783921      14340\n",
       "70                       SOMERSET  190042.943176      27498\n",
       "71                 WORCESTERSHIRE  190473.349404      26388\n",
       "72                  HEREFORDSHIRE  191937.252028       7669\n",
       "73                NORTH YORKSHIRE  192596.686729      28848\n",
       "74          SOUTH GLOUCESTERSHIRE  194889.125019      12794\n",
       "75                       CORNWALL  199168.369268      27400\n",
       "76                        BEDFORD  199342.258983       7971\n",
       "77                           YORK  199457.010277      10186\n",
       "78                  MONMOUTHSHIRE  200824.646281       3956\n",
       "79                 NORTH SOMERSET  201204.299395      11683\n",
       "80                    BOURNEMOUTH  202581.476247      10514\n",
       "81                SOUTHEND-ON-SEA  202829.761767       9059\n",
       "82                  CHESHIRE EAST  203600.412792      17859\n",
       "83                CITY OF BRISTOL  204345.825712      21591\n",
       "84                   WARWICKSHIRE  205537.558500      25625\n",
       "85                         SLOUGH  207672.811222       4803\n",
       "86           CENTRAL BEDFORDSHIRE  208143.014559      14946\n",
       "87                  MILTON KEYNES  210914.827990      12499\n",
       "88                          DEVON  212786.196210      41792\n",
       "89                GLOUCESTERSHIRE  213038.633608      31422\n",
       "90                      WILTSHIRE  217053.701071      23671\n",
       "91                           KENT  222816.214665      73717\n",
       "92                 CAMBRIDGESHIRE  226680.603347      32160\n",
       "93                    EAST SUSSEX  229368.771959      31054\n",
       "94                          ESSEX  234025.117029      70874\n",
       "95                        RUTLAND  235240.240618       1957\n",
       "96                         DORSET  239313.188820      22762\n",
       "97                        READING  243440.568329       7985\n",
       "98                      HAMPSHIRE  250628.699991      67579\n",
       "99                    WEST SUSSEX  255318.828667      45916\n",
       "100              BRACKNELL FOREST  264293.307341       6143\n",
       "101                         POOLE  265033.774489       8114\n",
       "102             BRIGHTON AND HOVE  276522.260494      14962\n",
       "103  BATH AND NORTH EAST SOMERSET  276606.562034       9021\n",
       "104                WEST BERKSHIRE  281778.408823       7781\n",
       "105               ISLES OF SCILLY  282605.156667         45\n",
       "106                   OXFORDSHIRE  295648.926991      30887\n",
       "107                     WOKINGHAM  306010.842833       8156\n",
       "108                 HERTFORDSHIRE  306911.661768      56536\n",
       "109               BUCKINGHAMSHIRE  333595.258420      26067\n",
       "110                        SURREY  386774.070890      61605\n",
       "111        WINDSOR AND MAIDENHEAD  442982.051869       7118\n",
       "112                GREATER LONDON  452225.698396     337924"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_sum_county = df.groupby('County').agg({'Price_adj': 'mean', 'count_col': 'count'}).sort_values(by='Price_adj').reset_index()\n",
    "pd.set_option('display.max_rows', None)\n",
    "df_sum_county"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Model"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Greater London"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x1dca6f31188>]"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "bins = np.arange(0, 1.2e6, 1e3)\n",
    "df_set1 = df_set1.dropna()\n",
    "df_set1_noout = df_set1[df_set1['Price_adj'] < df_set1['Price_adj'].quantile(.95)]\n",
    "y, x = np.histogram(df_set1_noout['Price_adj'], bins=bins, density=True)\n",
    "alpha, x0, inv_beta = gamma.fit(df_set1_noout['Price_adj'])\n",
    "\n",
    "# Plot of fit\n",
    "fig, ax = plt.subplots()\n",
    "ax2 = ax.twinx()\n",
    "ax.plot(bins[1:], y)\n",
    "ax.plot(bins, gamma.pdf(bins, alpha, loc=x0, scale=inv_beta), color='r')\n",
    "#plt.ylim([0, 3e-6])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Greater Manchester & Midlands"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(0.0, 4.2e-06)"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAEQCAYAAACgBo8fAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/Il7ecAAAACXBIWXMAAAsTAAALEwEAmpwYAAA6l0lEQVR4nO3dd3xUVfr48c8zk0roEGqA0ItUYWmiIlgoCosV66ooCzbUXde66Npddf2quLoo/lYF7A1ELCui0gnSESGBAKGGAAkldXJ+f9xJMplMkgmZzGRmnvfrNa/M3HvmznNJeHJy7rnPEWMMSimlgp8t0AEopZTyDU3oSikVIjShK6VUiNCErpRSIUITulJKhQhN6EopFSICmtBF5G0ROSQim3x0vLYi8p2I/CYiW0Qk0RfHVUqpYBDoHvp/gVE+PN67wPPGmO7AQOCQD4+tlFK1WkATujHmZ+CI6zYR6Sgi34jIGhH5RUS6eXMsEekBRBhjvnce+4Qx5pTvo1ZKqdop0D10T2YCdxpj+gN/Bf7t5fu6AMdE5DMRWSsiz4uIvcaiVEqpWiYi0AG4EpG6wFDgYxEp2hzt3Hcp8LiHt+01xlyEdS5nA/2A3cCHwI3ArJqNWimlaodaldCx/mI4Zozp677DGPMZ8FkF700D1hljdgCIyBfAYDShK6XCRK0acjHGZAE7ReQKALH08fLtq4GGIhLvfD0C2FIDYSqlVK0U6GmL7wPLga4ikiYik4BrgUkish7YDIz35ljGGAfWmPsPIrIREODNmolcKaVqH9HyuUopFRpq1ZCLUkqp0xewi6JNmzY1iYmJgfp4pZQKSmvWrDlsjIn3tC9gCT0xMZGkpKRAfbxSSgUlEdlV3j4dclFKqRChCV0ppUKEJnQP0o/nMuSZH0g+dDzQoSillNc0oXvw/ZaD7M/M4a1fdgY6FKWU8pomdA9szjIyhTpHXykVRDShe2BzFgbTfK6UCiZeJ3QRsTvL0n7lYV+0iHwoIskisjLYVwqS4h56YONQSqmqqEoPfRrwWzn7JgFHjTGdgJeA56obWCCV9NA1oyulgodXCV1EEoCxwFvlNBkPvON8/gkwUlwKmgcb0TF0pVQQ8raH/n/A34DCcva3BvYAGGMKgEygiXsjEZksIkkikpSenl71aP2kqIf+xbp9AY5EKaW8V2lCF5GLgUPGmDXV/TBjzExjzABjzID4eI+lCGqF4P3bQikVzrzpoZ8FjBORVOADYISIzHZrsxdoAyAiEUADIMOHcfqVTTO6UioIVZrQjTEPGmMSjDGJwERgkTHmOrdm84A/OZ9f7mwTtAPQdpsmdKVU8Dntaosi8jiQZIyZh7Vu53sikgwcwUr8QUvTuVIqGFUpoRtjFgOLnc+nu2zPAa7wZWCB5K8JOvuOZdOsXjQRdr2/SylVfZpJPPDHiMuRk3kMfXYRTy4ob2q/UkpVjSZ0D/xxUTQzOx+Axb8fqvHPUkqFB03oHugkF6VUMNKE7oHr/Jx9x7Jr9rNq9OhKqXCiCd0D11v+X/p+WwAjUUop72lC98C115xTUF61A6WUql00oXvgek/U/PX7amTYpegzdLheKeUrmtA9cK+D/uSCLTXwGTp6rpTyLU3oHrjn2rwC3ydfh47kKKV8TBO6B+6959wCh88/w+H8MyA14xQb0zJ9fnylVPjRhO6Be0L/ZfvhGv2MS2Ys8fnxlVLhRxN6BVxvMJr039UkPrDAZ8d26IKlSikf04TuQaGHGSg/bPXtLfoOvSiqlPIxTegeFOXamqy6WKg9dKWUj2lC96Ao19Zk1cUCTehKKR/ThO5B8ZCL9tCVUkHEm0WiY0RklYisF5HNIvIPD21uFJF0EVnnfNxSM+H6SdGQSw1+hI6hK6V8zZsVi3KBEcaYEyISCSwRkYXGmBVu7T40xtzh+xD9r6iHXpN10XWWi1LK17xZJNoYY044X0Y6HyGdjfYcPQVAhL1sQl+355hPPkNv/VdK+ZpXY+giYheRdcAh4HtjzEoPzS4TkQ0i8omItPFlkP60LPkwr/2YAsCbNwzgiv4JpfY/5aO6Lnrrv1LK17xK6MYYhzGmL5AADBSRnm5N5gOJxpjewPfAO56OIyKTRSRJRJLS09OrEXbN2bI/q/h5h/g4npxQ+lTzKimna4xh1c4jpSo2eqJDLkopX6vSLBdjzDHgR2CU2/YMY0yu8+VbQP9y3j/TGDPAGDMgPj7+NMKtea552CZClL30P1FOfsUJ/dvNB7nyP8uZs3J3he3ch1zSnMM8Sil1uryZ5RIvIg2dz2OBC4Ctbm1aurwcBwTtUvauidYmUmbqYk4lhbqKEvOO9JNefw7AIh/fiaqUCj/ezHJpCbwjInasXwAfGWO+EpHHgSRjzDzgLhEZBxQAR4AbayrgmuaaZj3NccnOqzihF/0CMJVcN3YfkdFrpEqp6qo0oRtjNgD9PGyf7vL8QeBB34YWGO49dHc5+ZUkdOfXyhK0ew+9sjF3pZSqjN4p6sY1r4rzX6dD07jibVk5BXy4uvzx8aLfAVVN0JrOlVLVpQndjadE7D4f/eOktHLfbysecqmYew9dJ70opapLE7ob1/nhRTk3wlb6n8leQdWuol2V3ThU6DZZRodclFLVpQndzfGc/OLnDWIjgbI99IoSetGYS2X52X235nOlVHVpQndz8HhupW3sNuFkbgH7M7PL7Cu+KFrJMcpcFNVRdKVUNWlCd7PniDWP/K6RncttE2ETrnlzBUOeWVRmn7cXRd336xi6Uqq6NKG7STl0gj8Nace9F3Qp3uaem+02YX1apnOf4dDxHA6fsHr2RRdF3cfI3ek8dKWUr3lzY1HYyHcUcjy3gKZ1o0ttv/IPbdi4N7P4dZ6jJPvmFhQy8KkfAJh76yAe/GwjUPkQinuPXKsvKqWqS3voLk457wKtE13699x1g9qWer3vWMnY+fGcguLn17xZUoRSbyxSSvmbJnRgY1omJ3ILuPKN5QDERdlL7RcRRvdsQbsmdQBIPnSieF+Wy6wYV5WlZ/f9+Q5N6Eqp6gn7hJ6d5+CSGUu4bc6v/H7wOFC2hw7w+nX9WfzX4WW2/2O+5/rolQ2huPfI8x2FnPnE98ytpEqjUkqVJ+wTelFtlnW7jxZvc++hF/G0aPTP28qp617ZPHS3/QcyczhyMo+HPt/Ioaycit+slFIehH1Cz3dOR4l0qXse56GHXlVfb9rProzyS+i69+A/W7u3+PnEme7LtSqlVOXCPqEXrUDkejdoXFT1E3pOfiEX/OvncvdXNO98x2HrF8HujFMkPrCApcmHqx2PUir0hX1Cz3UmdNdSuXWiPQ+5VFVeBQuHLv7dWtBi1p8GlNtmxc4MAD77dW+5bZRSqkjYz0PPdS4ptz+zZNy6Kj30CEcBo39fyjk719Lu2D4cNjvbmrZlUceB/JLYt9z3/bLd6nUP6tCk/IM7e/Eehu6VUqoMTegelpTztoc+ZNd6/rnwFdpkHiQjtj7bm7YlqiCfyzYt4k+/LiClcQIMj4AxY8o9Rnm5+st1e4vH9yuqBaaUUkUqTegiEgP8DEQ7239ijHnUrU008C7W4tAZwFXGmFSfR1sDioZcXMVGVp7Qb179JdMXvUlK49bcfNl0fuw4AONcESOqIJ8Lty/nniVzYOxYmDoV/vUviIkpcxxPqyIBTPtgXfG89/LaKKWUK2/G0HOBEcaYPkBfYJSIDHZrMwk4aozpBLwEPOfTKGtQnoeE7jrjxZNJqz5n+qI3+brLUMbe+DKLOg1kzq1DSo4ZEclX3c9h1M0z4L774PXXYeRIOHq0zLEqytXpzsqPms+VUt6oNKEbS9GtkZHOh/scjfHAO87nnwAjxdOk7VrIvYceHVHxP8myXqd4+Me3WdD1LO4cfz85kVavu2uLeiQ6e9RF8u2R8M9/wkcfwerVMHw4ZGSUaiMC7948sPh1lMvnl/TMra87D5/kyMm8qpyeUiqMeDXLRUTsIrIOOAR8b4xZ6dakNbAHwBhTAGQCZa72ichkEUkSkaT09HJuyPEz95rmFd7geeAAre78M1uad+AvY+/BYSsZmomw2fjRw52kAFxxBXz1Ffz+O1x8MZwsmZ9uE+GcLvGc370ZAPVc5sAX5fOiMfTzXljMeS8s9vbUlFJhxquEboxxGGP6AgnAQBHpeTofZoyZaYwZYIwZEB8ffzqH8KlvNu1n+pebS21zlJfRjYHbboNTp7jrkvuKe+ZF7HbxeCfp2f9cxLaDx+HCC2HuXMyqVfyvz3nYCq2LsUXvaNkgFih9U1NRD931sJnZnmvHKKVUleahG2OOAT8Co9x27QXaAIhIBNAA6+JorbZuT2aZbY7y7vj59lv4/HP4xz/Y0SQBgP9c3794d0Q5U1H2HMlm5s87rBeXXsr+x5/j/JTV3LXsA6AkaRe9vY6HsgN6UVQp5Y1KE7qIxItIQ+fzWOACYKtbs3nAn5zPLwcWmSCoB1u0ZmilCgvh/vuhQwe4557izRed0aL4eUXrjLqOix+87mY+6TmSu5e+z3kpq0uGVTy8v6g3rgldKeUNb3roLYEfRWQDsBprDP0rEXlcRMY528wCmohIMnAv8EDNhOtbUW4XQOtE2Zlzy6CyDefOhQ0b4MknISrK47HsFSTduSt3c8B541JmTgEPX3gbm5t14KWvXkT27wdKkvaJ3IJyj6OUUhXxZpbLBmNMP2NMb2NMT2PM487t040x85zPc4wxVxhjOhljBhpjdtR04L7gcFsn7te/X8BZnZqWblRYCE89BX36wFVXlXusoh72Ped38bh/8DM/MGflLrJyCsiNjOaO8fcTXZAPkyaBMcVDLhP6tS7zXu2gK6W8EdZ3irrPQY/xdEPRN9/A1q0wezbYKv+DZtr5nakTZeepr38rs+/hzzcVP9/ZuDVPn3cTT3zzBsycia3tOQDU8VB2QIdclFLeCOviXK7rhJbrxRehdWu48kqPu6df3KP4js4io3u18NjW3ex+Y+CCC+Dee5nYxBpqGd2z7Htr/9UIpVRtELY99Ow8B99uPlhxo02bYNEieO45iPR8AfXmYe25eVj7UtsSGtWhSVwUGZXcBGTEBrNmQY8etH/sflIXLPA4vvL20p0UFJZfuVEppSCMe+hFKxVV6O23rUR+882lNn979zksfWBEhW+deUP/CvcXa9MGnngCFi6ETz8F4G+jupZp9u7yXd4dTykVtsI2oRe4zTcvs+xcfr41bn7JJdC09IXSri3q0bphbIXH79umkffB3HEH9O0L06ZBVpbHC6ON6ng5xVIpFbbCOKGXHsJY8/cLSjdYsADS0+Gmm07r+BXNSy8jIgL+8x/Yvx+mT/d4EdR9iqVSSrkL2yxR4CjdQy8zw+Wdd6BFCxjlflOs976/5xzvGw8cCJMnw4wZRGz/vczuCC9m2CilwlvYZolyb/EHOH7cGtO+4gqr93yaOjWrW7U3PPEExMVR7+EHy+zaeyzbwxuUUqpE2CZ09zH0Ur7+GnJz4fLLq/UZVa4gHB8PjzxC1LcLGbZzbbU+WykVfsI4oZeMoZeZ+/3pp9C8OZx1Vo18dveW9cvfedddONq355FFb2Ev9GImjlJKOYVtQi+6S3Tm9f15/TqXKYanTlkXRC+9FOzerS1aVU9P6MmMa/px9cA2ZXdGR5Pz1DN0O7yLqzZ8VyOfr5QKTWGb0MfNWApAhN1tWOSHH6ykfumlNfbZ9WIiubh3K565tLfH/YV/nMDKhDO455c5xObl1FgcSqnQErYJvYjdffbIN99AXBycfbbPPqN907hSr+vHVnyh1Waz8dzwG4k/dYyb1szjkylDKmyvlFKgCb30whTGWLNbRo6E6GifHH/xX4fzxe2lx+Lrx1R8k5DdJvzaujvfdxrIlJWf0g7tpSulKqcJ3TWhb98OO3dWa+65u8SmcWUW0vBY1dFF0eSYF865gbq5p4h75SWfxaOUCl1hmdALXaYslhpDX7jQ+urDhH46iu4U/T0+kS97nEvsG68Rf+JI8X6v6tAopcJOWCb0bJeEWGoM/ZtvoGtXaN/ew7t8Y0S3ZpW2cb31/6Vh10J+fvEapABPLShba10ppbxZU7SNiPwoIltEZLOITPPQZriIZIrIOudjes2E6xtFa3UCxBYNf+TkwOLFcNFFNfrZt5/XsdI2rqNAuxu1pODmSUxc/y1tj1rL1aVmnKyp8JRSQcybHnoB8BdjTA9gMHC7iPTw0O4XY0xf5+Nxn0bpYxknrDrlf+zbii7Nnbfnr1plJfWRI2vkM68Z1BaABrGe1yR15X6HacFDj1Bgi2Da0rlAFQt/KaXChjdriu43xvzqfH4c+A0oW981iBw+kQvA9UMSS5LnTz9ZVyN9OF3R1aOX9ODjKUOqXN9l6xOjsLVuybtnjuWPW34i8cheLdSllPKoSplBRBKBfsBKD7uHiMh6EVkoImeU8/7JIpIkIknp6elVj9ZHTuRay73Vi3GZD/7TT9C7NzSqQh3zKoiOsPOHxMZVfl9MpB27CDMHXkqePZI7l39IpPvNUEopRRUSuojUBT4F7jbGZLnt/hVoZ4zpA7wKfOHpGMaYmcaYAcaYAfHx8acZcvXlO6zb/qPsztPPy4Nly+DccwMWU0XsNiEjrqHVS9+8mOYH9wQ6JKVULeRVQheRSKxkPscY85n7fmNMljHmhPP510CkiDR1b1dbFCX0yKJFI5KSIDu71iX0bi3qASVj6m8OnECePZILP38rkGEppWqpSot9i5VNZgG/GWP+VU6bFsBBY4wRkYFYvygyfBqpD+U5F7coHrr46Sfr6zlVWJCihn13zzk0rx9TatvhuEa8128Mtyz70roJqnPnAEWnlKqNvOmhnwVcD4xwmZY4RkSmiMgUZ5vLgU0ish54BZhojKmg4Hhg5TsrLUYXVVP8+Wfo0aPM2qGB1KV5vTJ3mALMHHQpjqgoePLJAESllKrNKu2hG2OWABVehTPGzABm+CqomlYy5CJQWAgrVsCVVwY4Ku8cjmvE+kuuYcDs/8Ijj2gvXSlVLCznvyUfOgFApN0G27bBsWMweHBgg6qClZfdZBUP0166UspFWCb0j9ekAc7CXCtWWBsHDQpgRN6LirCR1aAJ3HYbzJ5tjaUrpRRhmNCL5qCDc/bIypVQvz506xbAqLwXYROruNh991m99CeeCHRISqlaIuwSepZLHRfA6qEPGgRBcPelCNhFrAWumzeHqVNhzhztpSulgDBM6KVKz548CRs2BMX4+adTh7Lk/hHY7YKjqPzvffdBVBQ89VRgg1NK1QphmNCtGS7XDW4La9ZYs1yCYPy8f7tGtG4YS4TNJaG3aAFTplhj6SkpgQ1QKRVwYZfQcwusHvrI7s2D7oIoWLXSHS4LdPC3v0FkJNmPPc6Rk3mBC0wpFXBhl9CLeujRETbrlv/27WvVDUWVibBZY+j7jmVbib1lS5g8mcg5sxn/l/cCHZ5SKoDCL6E7e+gxkXZYuxbOPDPAEVWN3S7sz8xm6LOL6PjQ13yUtAfuvx+Hzc7tyz9id8apQIeolAqQsEvouc4eeuyp45CcDP36BTiiqrGLcCgrt/j1O8tSoVUr3u9zEZdt+oEfv/VU2VgpFQ7CLqEXzXKpv22LtSHYeug2odClTM7mfVkUFhreGHQ5hSL0m/16AKNTSgVS2CX0ohuL6m/ZaG0Ish56hM2Ge9Wz9WnHOFC/KR/2vogzvvscdu0KSGxKqcAKq4Sek+/gkS82ARCzaYM17a9FiwBHVTU2m+Bex7LoQu/rgy/HCPDMM/4PTCkVcGGV0IuKcgFErF8XdMMtYM1yyTiRW2rb5n2ZAOyvH88HPS+g8O23YffuQISnlAqgsEroUc4ViqLzc5Hffgu64RaA9OO5ZOUUlNr25ILfip//e8gVVg/+2Wf9HJlSKtDCKqEXDVV0S08FhyMoe+gHsnIq3L+vfjOSx14Os2ZBWpqfolJK1QZhldDznCsVnXFoh7UhCHvotgqXGrFsu+l2TGGh9tKVCjOVJnQRaSMiP4rIFhHZLCLTPLQREXlFRJJFZIOI1Mqub55zpaIzDqZAw4aQmBjQeE5H0YLRFZmxw8EHPUbgmPkm7N3rh6iUUrWBNz30AuAvxpgewGDgdhHp4dZmNNDZ+ZgM1MrJ0EU99NFkQK9eVj3aIFOqjks5fj94nNeGXGkVHnvuOT9EpZSqDSpN6MaY/caYX53PjwO/Aa3dmo0H3jWWFUBDEWnp82irKd9RCMZQP+V36Nkz0OHUqLSGLdg0chzMnAn79gU6HKWUH1RpDF1EEoF+gPv95a2BPS6v0yib9BGRySKSJCJJ6enpVQy1+vIKCmlxPIOI41khn9ABVl49BQoK4J//DHQoSik/8Dqhi0hd4FPgbmNM1ul8mDFmpjFmgDFmQHx8/OkcolryHYV0Oey8izIEEvqFPZrz+W1D+b+r+nrcf6JVW7j+evjPf2D/fv8Gp5TyO68SuohEYiXzOcaYzzw02Qu0cXmd4NxWq+S5JvQzzghsMD5w49BE+rVtROfmdT3ut9kEHn4Y8vPh+ef9HJ1Syt+8meUiwCzgN2PMv8ppNg+4wTnbZTCQaYypdV3C4zkFdE3fjaNFC2jSJNDhVFvR9dFIu+dvo10EOnWCa6+FN96Agwf9GJ1Syt+86aGfBVwPjBCRdc7HGBGZIiJTnG2+BnYAycCbwG01E271ZGbnWz30EBhuAYqrLpaX0F/8fhsrdmTAI49Abq720pUKcRGVNTDGLAEqnN9njDHA7b4KqqYcO5FDl8O7sfe6ONCh+ERRQo+o4G6jiTNXkPrsWLjmGvj3v60l65o181eISik/Cqs7Re2pqcQW5AZ1D/2rO4cVPzeVDLkU+ShpT0kv/YUXajI8pVQAhVVCr5vyu/UkiBN6z9YNOK+rNUOoqIdur6QewN8+2UBK49YwcSK89hoEYMqoUqrmhVVCr7d9q/Wkh/uNrsHlj/2sKf5dmtcDShJ6ozqRpD47lvWPXljmPQczc6xeenY2vPii/4JVSvlN2CT0zFP5NNq5jd0NmkNdz9P8gsX4vq3Z+cwY2jSuA0CE3UroTetGW6899NgPn8yD7t2tXvqrr+qMF6VCUNgk9KOn8uiSvoujiZ0DHYpPuBbpqh8TyYtX9OG9SYMAz0Mwd72/lsQHFrDm5mnWWPpTT/ktVqWUf4RNQi/Iy6P90b3E9esd6FBqxGX9E2jRIAaoeNbLo7/lwc03W/PSU1P9FJ1Syh/CJqGbnalEOwrI7tAp0KHUuIouku7OOAXTp4PNBo895r+glFI1LmwSun37NgByO4bGkEtFKqqZHhVhh4QEuOMOeO892LLFj5EppWpS2CT0yOTwSegVibILTy3YwrvnTIS4OGvmi1IqJIRNQo9K3k5GbH2kafDXcPHG13ed7XG7zSa8+ctOpi87CH/9K3z+Oaxa5efolFI1IWwSekzKdlKaJBBVyV2VoaJHq/q8cd2Z/POy0heB853L8AFwzz0QHw8PPeTn6JRSNSE8shsQu2M7KY0TiAiThA4wqmdLrvxDG2ZPGsTfRnUF4GBWbvH+hakn2DP1bvjhB+uhlApq4ZHdjhwh6kgGOxonEGkPvnVEq2tY56bcNrwTQzqUHm6aOudXzj/ZHdq0gQcfLCkOo5QKSuGR0H+3arikNEmotJBVKDuQlVNmW25EFDz+OKxeDR9+GIColFK+EhbZzWy1arjsaNw6rBP6zsMnPe+4/nro0wceeAByyiZ9pVRwCIvslp60njxbBHsatiA20h7ocGofu90qq7trl1XnRSkVlMIiocekJJPaqBUOm534etGBDqd2Ov98GDPGqvFy+HCgo1FKnQZv1hR9W0QOicimcvYPF5FMl+Xppvs+zOqJTtnOjiZWydnKaoeHsrvPr+SmquefhxMn4B//8E9ASimf8qaH/l9gVCVtfjHG9HU+Hq9+WD6Un09U6g5SGidw49DEQEcTUHef38Xj9u0Hj1tPevSAW2+1Cnc5LyQrpYJHpQndGPMzcMQPsdSMHTuQggJSmiQw3LnSjyrtgpd+5tvNBzDGWL3z2Fhr7VGlVFDx1Rj6EBFZLyILReSM8hqJyGQRSRKRpHR/LYPm7Glac9DD4pJBhVo3jPW4/c/vreH7LQetBaQffBDmzYMff/RzdEqp6vBFhvsVaGeM6QO8CnxRXkNjzExjzABjzID4eD/1lrdZRbl2NG5dYZ3wcPHtPeew6qGRzL9jWJl9ry5KxlFo4O67oW1bmDYNCgr8H6RS6rRUO6EbY7KMMSecz78GIkWkabUj85XkZPIbNCIrpm5Y3fZfnrrRETSrH0OvhAa8feOAUvs27s1k/GtLSD1ZCC+9BBs3wuuvByhSpVRVVTvDiUgLcRbgFpGBzmNmVPe4PpOSwqm2iQBhedt/RYZ3aVZm26a9WVw1czlMmAAXXAB//zscOhSA6JRSVeXNtMX3geVAVxFJE5FJIjJFRKY4m1wObBKR9cArwERjalFRkORkTrZJBCDCpj10V7ZyhqAOZuWCCLzyCpw8aY2pK6VqvYjKGhhjrq5k/wxghs8i8qW8PMzu3XzcajCgPfSqWLT1IMdO1eWP0+7G9uILMHkyDBoU6LCUUhUI7S5raipSWMiuRi0BdAy9Cp5c8Bv3frSeN4dfA61awe23g8MR6LCUUhUI7QyXkgLAroZWQs8rKKyotXJxMtea3ZKaF2HVeVmzBmbNCnBUSqmKhHZCT04GKO6hd4iPC2Q0QaVoIYwCRyFMnAjnnmtVY9QLpErVWqGd0FNScNSJ43Cdhgxq31hvLPLgprMSK9yf7yjk4zVpJP/jeesC6d13+yUupVTVhXaGS04mp117EGFaZYWpwtSjl5zB+kcvLHd/vsNw3ycbOH/hIWvt0fffh4UL/RihUspbIZ/Qs9u0AwibxaFPR3RE+f82CzbuL3nxwAPQvTtMnWpVZVRK1Sqhm+UcDti5k5Nt2gM6w6UiMZF2Vjw4svKG0dEwc6a1EMb0WlclWamwF7pZLi0N8vI4kdAW0DnolWnRIMa7hsOGwZQp8PLLkJRUs0EppaokdBO6c8piVmsdcvHWtJFeXmd49llo3hxuuQXy8mo2KKWU10I3yzmnLGa1TgR0yMUb91zQhdRnx/LK1f0qbtiggVW0a/16ePJJ/wSnlKpU6Ga5lBSIiiKriVWASodcvHfslOde92PzNpe8GD8ebrgB8/TTvPWvD0k+dNxP0SmlyhO6CT05GTp0IMdYiTyqgpkcqrRzOnuuVf/fZanszjhVsuHll3E0b8Hwp/7KHbOW+Sk6pVR5QjfLpaRAx44cO2n1NhvGRgU4oOCR2DSu3F+A5zxvrWK0PzObdcfh4Euv0elIGjct1LIASgVaaCZ0Y6weeqdOZJzMo0FspPbQq+izqUPL3ZeT72D484v542tLOTzkXN7rN4YrlnwCP//sxwiVUu5CM8sdOmTdpt6xI4dP5NIkTnvnVdWzdYNy93X7+zfkOgudHc8p4JnhN3GgUUu44QY4dsxPESql3IVmQnfOcKFTJzJO5NGkrib001F0IblRnchy22Rm53MqKpanrn0Y9u6FW2+1/kJSSvldaCZ05xx0OnYk42QuTeKiAxtPkCoqZmavYHHtg1k5ACyo0w7zxBPwySfw5pt+iU8pVZo3S9C9LSKHRGRTOftFRF4RkWQR2SAiZ/o+zCpKTgabDRITtYdeDUUJvajDfbeHAmefr91b/PyRzqOtdUinTYPNm8u0VUrVLG966P8FRlWwfzTQ2fmYDAR+mfiUFGjbFkdEJEdO5dGkrvbQT8dlZyYA8MDobgB0aV6vTJuNezOLn89Znca7Ux7H1K8PV10F2dn+CVQpBXiR0I0xPwNHKmgyHnjXWFYADUWkpa8CPC3OGS5HT+VhDDTVHvppeXhsd9Y/eiFXDGhD6rNj6VXBhdIi01dlcOfoe6we+p13+iFKpVQRX4yhtwb2uLxOc24rQ0Qmi0iSiCSlp6f74KPLUZTQnXPQG9XRhH467DahQWzJBdE2jet49b6vWvRi0WW3WkvW6Xi6Un7j14uixpiZxpgBxpgB8fGe70astqNH4cgR6NiRk3nWosZx0faa+awwNP+OYR63PzSmW6nXt3S4mB39h8Edd5C7dBmbXIZmlFI1wxcJfS/QxuV1gnNbYBTNcOnUiWxnQo+NjAhYOKGmV4LnYZcbhiSWel1oszPhrNvJbNyM7HETuPHZ+Rw+keuHCJUKX75I6POAG5yzXQYDmcaY/ZW9qca4TFnMzrdWro+N0h56TYuJLPtvnBlbj6tH/Y2YrGPM+PI5Tp3UhK5UTfJm2uL7wHKgq4ikicgkEZkiIlOcTb4GdgDJwJvAbTUWrTeKbirq0IHsPOtuxjqa0H1uYGJj5t46qNJ2W5p34MGL7mDwnk00eug+velIqRpU6ViEMebqSvYb4HafRVRdKSnQsiXExXEqz5qcE+uh96hO37YnR2O3SZkbjqYO78jri1PKtP+85wi6pqcy5e2Z0LuHNU9dKeVzoXenqHOGC0B2vnMMXXvoPhUVYfN49+htwzuW+57nht/IybHj4J57YP78mgxPqbAVegndWTYX4Lf9WQDUi9GLov5QUUVLIzb2vToT+veHq6+GtWv9GJlS4SG0EvqpU7BvH3TqRG6Bg/dXWdPjoyO0h+4P0RF23rphQLn7F+7IgnnzoHFjuOQS2L3bj9EpFfpCK6Hv2GF97diRHeknAxtLmPhg8mAWTju7+PX5PZrz698v8Nj2X99v40CdRrBgAZw4ARdeCDV5g5lSYSa0ErpL2dwc5/i5qlmDOzShe8v6pbY1josqdw3XaR+shV694KuvrB76qFGQleWPUJUKeaGV0F3moBctwPDYJT0CGFD4atPIc5mAlTudZYGGDbNK7W7YYC04nZPjx+iUCk2hldCTk63x2UaNyHMm9PLubFQ1a86tgzi7c9OKG40ZA++8Az/9BFdeCXl5/glOqRAVWgk9JaV4ymJRD10viAZGywaxvDdpEE9N6FlmX+IDC5jy3hrrxTXXwGuvWVMZr7hCk7pS1RBaCT05uXjKYm6BNYYerYtDB9S1g9rx+rVl1zz5ZvOBkhdTp8KMGdYMmMsvh1wtEaDU6QidbJeXB7t2FSf0PO2h1xoNylmT9Pa5v3Iit4C8gkJ+PP8KHrnwNqunftllOqau1GkInTtudu2CwkLobC2TtveotVpORTe7KP8Y2rEp/++mP/D64hRW7SxZK2XBhv0s2OBSx63fGC7ul8Dg5x6icNw4bJ99BnXrBiBipYJT6GQ7lymLAC9+vw3QOi61xXldmzHxD20qbbf9j9dw3+hp8L8fYMQInaeuVBWEbELv1sJa/7K8P/eV/116ZgIpT4+psM2qnUf4uPcFTJ7wMDlr12OGDbP++lJKVSq0Enq9euBcCalxXBR/SGwU4KCUO09FvVzNX78PgP91HsR1Vz2BOXgIhg615qsrpSoUWgm9UycQK2HkOwp1/DzIJSWcwajLniS/EDjrLPjyy0CHpFStFjoZz6Vs7sGsHFanHiXlkNZzqY3c1yX9YPJgVjw4ko/+PKRM223xifz72dnQrRtMmADPPKOLZChVjtBI6AUFsHNn8ZTF77ccBOBAlk59q416JTQg9dmxLJx2NjOu6cfgDk1o0SCGge0bk/rsWIZ2bFKq/Zs78vji5blw1VXw0ENw3XWQnR2g6JWqvbxK6CIySkR+F5FkEXnAw/4bRSRdRNY5H7f4PtQK7NkD+fnFPfR8R6FfP16dnu4t63Nx71Zlto/o1qzU6xO5Bdw9bxvLHn+FjVPvg7lzYcgQ2LbNX6EqFRS8WVPUDrwGjAZ6AFeLiKeKVx8aY/o6H2/5OM6Kuc1wKXDon+TBbNKw9nw6dWiZsgHXzFrFJfXP5fgnn1u/xAcMgI8+ClCUStU+3vTQBwLJxpgdxpg84ANgfM2GVUVuCT1Pe+hBTUTo364R1w5qR9vGZas27hp8nrXiUc+ecNVVpE68kRWb9wQgUqVqF28SemvA9X9LmnObu8tEZIOIfCIiHu8gEZHJIpIkIknpvrxhJDkZYmOtxaGBwye0Fkio+N+95zLjmn6ltl386hKmr82i01n385+Bl5L44Ts0PWcorF7NoeM5ZLh9/zftzeTISS36pUKfry6KzgcSjTG9ge+Bdzw1MsbMNMYMMMYMiHfOF6+uJ77awvfzlloXRG3W6aQ5b/tvUT/GJ5+hAicqwkYdD4t8v7t8FwX2CJ4572auvepJ6uTnYIYM4f1RN3Hxi4tKtb341SVc8cYyf4WsVMB4k9D3Aq497gTntmLGmAxjTFG36C2gv2/Cq9ysJTtpd3R/8XALwImcAiLtwrw7z/JXGKoGDe/SjCf/2JOtT4zyuBLS0sS+jLp5Bp91P5dpyz7gzdfvZOyNLzPm5V+Ki7SlpJ8s03NXKtR4k9BXA51FpL2IRAETgXmuDUSkpcvLccBvvguxYmIKaXdsP4XOKYsA2fkOhnRsSrN62kMPBTabcN3gdsRE2lly/wiuG9y2TJusmLr8Zey9/HnCQzQ/kcG8d+/l8vde4OkPVhS3eeOnFH+GrZTfVZrQjTEFwB3At1iJ+iNjzGYReVxExjmb3SUim0VkPXAXcGNNBeyuxfEMoh35ODq4JPQ8B3W0KFdIal4/hifG96Rvm4ZA2VIC33YZyshb3mBO39HcuGY+U6ZewpitS8AYLaWsQp5XY+jGmK+NMV2MMR2NMU85t003xsxzPn/QGHOGMaaPMeY8Y8zWmgzaVeJRq/ZHQWKH4m2n8guI9TDuqkKDiPDp1KFsfWIUfxqSWGZ/Vkxdpl84lQnXv8DhuEb8+8tn+fD9B/l57tcUOArZdvA4xhgys/PJznOQW+Bg1c4jJD6wgEe+2Oj/E1LKR4K+HnrHI9Zwfl7nzsQ6t2XnFWpCD3F2m2C32fnbqK4YDJv3ZhFfL5oFG0vqq69v1ZXxN/yLq9d/y91L5jDv3Xv5IulLXjjnBtIaNC9u1zuhAc3qRQMwe8VuHrvkDCLsNtKP59IgNlJrAqmgEfwJPWMPJyNjyGtu3XF4MCuHwydydcglTMRE2nn0kjMAcBQaLunTih9+O8jHa9K4a2RnWtSP4SGbnS96DOfPKz/l1tWfM/r3ZcztO4o3Bl3GwXpN2ZCWWeqYnR5eyJ+GtOOd5bsY16cVr1zdz9NHK1XrhEBCTyOlSQLxzoJNKYdOANCzdYNAhqUCwG4TRvVswQU9mvPouDOoG239eHduXpf1e47xZHQd5vQdzd1L53Ld2q+5Zt1CPuhzEa8PuoID9ZuWOtY7y60a7PPW7+PMtg1JO5rN7JW7WDf9QmK0s6BqqaD/W7LDkTRSGicU3+6fmZ0PQFfnAhcq/NhtUpzMAf6Q2Jhbzu7A8gdHcKB+Ux4YfRd7lq/lf38YxTXrvuGnmbfwzMJXeK6b5/8Oj83fwltLdpKTX0i/x79n1c4jHNTCb6oWCu6EfvIkCVnppDRJKC7IdcyZ0BvE6kpFqrSWDWKZcU0/7h/VjQ4DezF2+Xzen/0DH/e6gCu2/sRVN43h16Uv8Fqjg3RvFufxGNn5Dq78z3IGPf0Dv+4+yiNfbOT6WSsBSD50nAn/Xsra3Uf5fG2aP09NKSDYh1yc1faSm7ShoNDqoRctDt1Ql55THrhXdzx/9EBejX6OnEHx1H3nbRrPmMHYByYxtls35va+iBeaD+RIHc/Dd5f+u+Tu0282HWDuqt2s3X2MCc7tl/RuRYQ9uPtMKrgE90/bVmt2ZErjkh7691sO0rdNQ+pEBffvKuUfrRrG8sylvajbuoVVaz01FWbPhgYNuOajl1nzxo18u+QVhqckYSt0lHucKbPX8PO20vWJLn51CVk5+Ww7eJwjJ/N45IuNLEs+TOap/FJDNoeO5/CP+ZuL72pV6nQFd9bbuhWH2NjVqBUFDsNv+7P4/eBxnWamTl9UFFx7rfXYvBmZNYuu773Hf5d+R2ajphw4/2K+7HY2r+fGY6Tin7OtB47T+7HvSm2bvWJ38fOUp8cgwOPzt/DVhv0UFhouPTOBvceyOXIyj+sGt6uJM1QhTEyAlvMaMGCASUpKqt5BrryS1P8tZfjkmXw6dQiHT+Tx5/fWcOPQRB4bd4ZvAlUqLw/mz4c5c+DrryE3l4LWCfy/VgNY2OUs1rXqQqGt6jNfhnVqypLkwxW2Ob97Mx4Y3Y3Hv/qNQe0b0zguiqsHWqUPChyFHod0jpzMo3FcVJXjUcFBRNYYYwZ43BfUCb1nTxbn1eXGS//OnFsG8coP21m58wiL/zqcxKaeL2opVS1ZWTBvHnz4Iebbb5H8fByNG7Ox5xDebtCDn9qfSWZsPe67qCuFhYYXv99GqwYx7Mu0hljuGtmZV37YXu0w7jivEzN+TGZMrxb85cKudIyvC8Bnv6Zx70fr+erOYZzRqj4iZYuZqeBWUUIP3iGXnBzYupXkYRMBOJlbwMqdRwC9IKpqUP361pqm112HHDsG332H/auv6LtwIa8cXoCx2cjt/wdiIs7nf817EJMfzXVDujJpWHt2pJ+kczMr8XpK6g1iI4un3VZmxo/Woi5fbzzA1xsP8MDobozt1ZIXv7MmClz86hIAXp7Yl0JjOJXn4KyOTWlaL5q60REYY/g4KY2xvVuyfs8x4utF07m5TvUNdsHbQ1+zBgYM4JFrpzM7YSAX927JVxus2753PD0Gm017JsqPHA5YvRoWLIDvvrN+Ph0OHBGRyKCB2EaMgKFDYeBAaNyYpNQjXP7GcsDqtd85ohORdhs/bUvn/ZW7efHKPmTnO7jwpZ99vjiH618MTetGFy8Ik/rsWJ9+jqoZoTnkMmsW3HILNzw4m58LGxZv/vvFPZg0rH31A1SqOrKyYMkSWLzYeqxZA4XOWSydOsHAgRT0/wMRQwdDnz7Wilse5DsK+W1/FuNmLC2z7/nLe9OzdQNGv/yLT0J+ekIvrhlUUpr43o/W8dmve/nm7rPp1qK+Tz5DVV9oDrmsXw9163K0eQLsP1G8eWjHJgEMSimn+vVhzBjrAVaCT0qCVausx08/ETF3rrVPxEryvXpZ66T26mU9OnYkMiKC3gkN2frEKGIi7fy/pTtZnXqE9k3juOzMBPILS6Y6LntgBP/8ZitPTejFybwCVuw4Ql5BIbsyTrJ5XxaLth6qMOSHPt/IQ59vpG50BKseHslnv1qF70b93y/aew8SwZvQ166F3r2JiioZL79hSDu6t9SehKqF6teHESOsR5G9e63kvn49bNwImzbBF1+U9OSjo62lFTt3JqZTJ+jcmZs6deKmsztDQgLYhGjn7JrhXeNp1TCW/5toFRKLi45gXJ/SN1GtTj3C3JW7+XxtyYJjrRvGsvdYdql2J3IL6DH9W9//G6gaF5xDLrm50LAh5s9/5sK2E9juLMi16qGRNNN1RFUwy86GLVusBL95M2zfbi2Cnpxs/dwXiY62knqbNhQmJCBt2iBt20KbNtYjIQEaNbJ6/5XIzM7n1neTWL/nGBee0YL56/eVabPzmTE6Y6aWCKkhl3xHIclf/Uj3nBym7Ipje0zJcEvdmKA7HaVKi42F/v2th6vCQqtH75rgd++GPXuw/fQT7NtnXZh1FRkJzZpB8+aev8bHQ6NGNGjUiI8mdLR+AURG8uylvdh64DiXvb6Mrs3r8fvB4+Tk6xoDwcCrDCgio4CXATvwljHmWbf90cC7WItDZwBXGWNSfRuqJeNEHl+++iHdgaSEHgDERNqsHzgta6pClc1W0vt2HbYpUlAABw7Anj3WIy0NDh2CgwdLvm7ebH3Nq2DWTFwccQ0b0r9RI1IbNiSNGFYedVCYMR8a1Ye6dUsecXGeX8fFQUyM9YiI8OqvBOUblSZ0EbEDrwEXAGnAahGZZ4zZ4tJsEnDUGNNJRCYCzwFX1UTALRrEMG7nSrY0a09GXEM6N6vLJ1OHkp3n0D8JVfiKiLCGWRISYMiQ8tsZY12gPXQI0tPh6FE4dsz66uF5vf37GLT/MNH71sOpk6WHfbwhUpLco6NLf/X0PDraOpfISOur63P3r5Vti4iwfhGW97DbfbNfpOyjvO1Fj6go61x9zJse+kAg2Rizw/r+yAfAeMA1oY8HHnM+/wSYISJiamKAfv58euz+jR1/e5TtT40m0nnrs5bLVcoLItCggfXo3LnS5puSD3PtWyuJibRhF8Fe6CA2L4c6ednE5udQx/ncep1b/DzKkU90QT5RBflEOfKIKsgnuiDP+Tqf6Ow8ok7kE1VwqmS7c19EoQN7oYOIwgLshYVEFDqIcBRY20xoFDBbdeWtDPxwps+PW+lFURG5HBhljLnF+fp6YJAx5g6XNpucbdKcr1OcbQ67HWsyMNn5sivw+2nG3RSouAhG8NBzqZ1C5VxC5TxAz6VIO2NMvKcdfr2KaIyZCVT715KIJJV3lTfY6LnUTqFyLqFyHqDn4g1v6szuBdq4vE5wbvPYRkQigAZYF0eVUkr5iTcJfTXQWUTai0gUMBGY59ZmHvAn5/PLgUU1Mn6ulFKqXJUOuRhjCkTkDuBbrGmLbxtjNovI40CSMWYeMAt4T0SSgSNYSb8m+f5qQuDoudROoXIuoXIeoOdSqYDdKaqUUsq3dK02pZQKEZrQlVIqRNTqhC4io0TkdxFJFpEHPOyPFpEPnftXikhiAML0ihfncq+IbBGRDSLyg4jU2hWCKzsXl3aXiYgRkVo51cyb8xCRK53fl80iMtffMXrLi5+vtiLyo4isdf6MjQlEnJURkbdF5JDz3hZP+0VEXnGe5wYROdPfMXrLi3O51nkOG0VkmYj0qfaHGmNq5QPrAmwK0AGIAtYDPdza3Aa84Xw+Efgw0HFX41zOA+o4n08N5nNxtqsH/AysAAYEOu7T/J50BtYCjZyvmwU67mqcy0xgqvN5DyA10HGXcy7nAGcCm8rZPwZYCAgwGFgZ6JircS5DXX62RvviXGpzD7245IAxJg8oKjngajzwjvP5J8BIqZ0FXSo9F2PMj8aYU86XK7Dm+9dG3nxfAJ7AqumT48/gqsCb87gVeM0YcxTAGFPxChGB4825GKBosYAGQNkaubWAMeZnrJly5RkPvGssK4CGItLSP9FVTWXnYoxZVvSzhY/+z9fmhN4a2OPyOs25zWMbY0wBkAnUxiWLvDkXV5OweiG1UaXn4vwzuI0xZoE/A6sib74nXYAuIrJURFY4q47WRt6cy2PAdSKSBnwN3Omf0Hyuqv+XgoVP/s9rAfFaRkSuAwYA5wY6ltMhIjbgX8CNAQ7FFyKwhl2GY/WefhaRXsaYY4EM6jRdDfzXGPOiiAzBum+kpzEhUu0qiInIeVgJfVh1j1Wbe+ihVHLAm3NBRM4HHgbGGWOqWKfUbyo7l3pAT2CxiKRijXPOq4UXRr35nqQB84wx+caYncA2rARf23hzLpOAjwCMMcuBGKwCUcHGq/9LwUJEegNvAeONMdXOXbU5oYdSyYFKz0VE+gH/wUrmtXWsFio5F2NMpjGmqTEm0RiTiDU2OM4Yc5rrDdYYb36+vsDqnSMiTbGGYHb4MUZveXMuu4GRACLSHSuhp/s1St+YB9zgnO0yGMg0xuwPdFCnQ0TaAp8B1xtjtvnkoIG+ElzJVeIxWL2iFOBh57bHsRIEWD+UHwPJwCqgQ6Bjrsa5/A84CKxzPuYFOubTPRe3touphbNcvPyeCNbw0RZgIzAx0DFX41x6AEuxZsCsAy4MdMzlnMf7wH4gH+svpEnAFGCKy/fkNed5bqytP1tenstbwFGX//NJ1f1MvfVfKaVCRG0eclFKKVUFmtCVUipEaEJXSqkQoQldKaVChCZ0pZTyg8qKdXloX+XCcDrLRSml/EBEzgFOYNWi6VlJ285YN4KNMMYcFZFmxov7U7SHrpRSfmA8FOsSkY4i8o2IrBGRX0Skm3PXaRWG04SulFKBMxO40xjTH/gr8G/n9tMqDKfFuZRSKgBEpC5WTfSPXap+Rzu/nlZhOE3oSikVGDbgmDGmr4d9aVgLXuQDO0WkqDDc6soOqJRSys+MMVlYyfoKKF5er2gZui84jcJwmtCVUsoPROR9YDnQVUTSRGQScC0wSUTWA5spWWnqWyBDRLYAPwL3GS/K6+q0RaWUChHaQ1dKqRChCV0ppUKEJnSllAoRmtCVUipEaEJXSqkQoQldKaVChCZ0pZQKEf8f4NDZg03dgwsAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "bins = np.arange(0, 1.2e6, 1e3)\n",
    "df_subset = df_subset.dropna()\n",
    "df_subset_noout = df_subset[df_subset['Price_adj'] < df_subset['Price_adj'].quantile(.95)]\n",
    "y, x = np.histogram(df_subset_noout['Price_adj'], bins=bins, density=True)\n",
    "alpha, x0, inv_beta = gamma.fit(df_subset_noout['Price_adj'])\n",
    "\n",
    "# Plot of fit\n",
    "fig, ax = plt.subplots()\n",
    "ax.plot(bins[1:], y)\n",
    "ax.plot(bins, gamma.pdf(bins, alpha, loc=x0, scale=inv_beta), color='r')\n",
    "plt.ylim([0, 4.2e-6])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## UK"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "bins = np.arange(0, 1.2e6, 1e3)\n",
    "fig, (ax1, ax2) = plt.subplots(1, 2)\n",
    "fig.set_figwidth(10)\n",
    "\n",
    "\n",
    "#=============================\n",
    "# Plot 1\n",
    "#=============================\n",
    "perc = df['Price_adj'].quantile(.95)\n",
    "df_noout = df[df['Price_adj'] < perc]\n",
    "#df_noout = df[df['Price_adj'] < 5e5]\n",
    "y, x = np.histogram(df_noout['Price_adj'], bins=bins, density=True)\n",
    "alpha, x0, inv_beta = gamma.fit(df_noout['Price_adj'])\n",
    "\n",
    "ax1.axis([0, 1.2e6, 0, 7e-6])\n",
    "ax1.plot(bins[1:], y)\n",
    "ax1.plot(bins, gamma.pdf(bins, alpha, loc=x0, scale=inv_beta), color='r')\n",
    "ax1.axvline(perc, color='g', linestyle='dashed')\n",
    "ax1.grid()\n",
    "\n",
    "plt.setp(ax1, xticks=[0, 2e5, 4e5, 6e5, 8e5, 1e6, 1.2e6], \n",
    "         xticklabels=['0', '200k', '400k', '600k', '800k', '1M', '1.2M'])\n",
    "plt.setp(ax1, xlabel='Price (£)')\n",
    "plt.setp(ax1, ylabel='pdf')\n",
    "ax1.legend(['Empirical distribution (p95)', 'Gamma fit', 'p95 threshold'])\n",
    "\n",
    "#=============================\n",
    "# Plot 2\n",
    "#=============================\n",
    "perc = df['Price_adj'].quantile(.99)\n",
    "df_noout = df[df['Price_adj'] < perc]\n",
    "y, x = np.histogram(df_noout['Price_adj'], bins=bins, density=True)\n",
    "alpha, x0, inv_beta = gamma.fit(df_noout['Price_adj'])\n",
    "\n",
    "\n",
    "ax2.axis([0, 1.2e6, 0, 7e-6])\n",
    "ax2.plot(bins[1:], y)\n",
    "ax2.plot(bins, gamma.pdf(bins, alpha, loc=x0, scale=inv_beta), color='r')\n",
    "ax2.axvline(perc, color='g', linestyle='dashed')\n",
    "ax2.grid()\n",
    "plt.setp(ax2, xticks=[0, 2e5, 4e5, 6e5, 8e5, 1e6, 1.2e6], \n",
    "         xticklabels=['0', '200k', '400k', '600k', '800k', '1M', '1.2M'])\n",
    "plt.setp(ax2, xlabel='Price(£)')\n",
    "plt.setp(ax2, ylabel='pdf')\n",
    "ax2.legend(['Empirical distribution (p99)', 'Gamma fit', 'p99 threshold'])\n",
    "\n",
    "plt.tight_layout()\n",
    "#plt.savefig('figures_mixture/gamma_fit.png', dpi=600)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "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": [
    "x = df['Price_adj']\n",
    "log_x = np.log(x)\n",
    "log_bins = np.arange(8, 16, 0.01)\n",
    "plt.hist(log_x, bins=log_bins, alpha=1)\n",
    "plt.xticks(np.log(10**np.array([3, 4, 5, 6, 7])), ['1k', '10k', '100k', '1M', '10M'])\n",
    "plt.show()\n",
    "\n",
    "# Gamma fit"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABCwAAAEUCAYAAADgA6f3AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/Il7ecAAAACXBIWXMAAAsTAAALEwEAmpwYAACnXElEQVR4nOzdd3hUVfrA8e+Z9GTSGwkJCQkQIHQQEERiQwFR7AVWsWF31XXVtay4lnVtv9XVXeuCBQu6NhBRQaMiSpPeQwkkIRVSJj2Z8/vjJpESSEhmcieT9/M880xy75173zOBMzPvnPMepbVGCCGEEEIIIYQQwpVYzA5ACCGEEEIIIYQQ4kiSsBBCCCGEEEIIIYTLkYSFEEIIIYQQQgghXI4kLIQQQgghhBBCCOFyJGEhhBBCCCGEEEIIlyMJCyGEEEIIIYQQQrgcSVgIIYQQQgghhBDC5UjCQgghhNMopW5TSq1SSlUrpeY46RqXK6W2KKXKlVI7lVLjnHEdIYRwBdKvCiG6Ek+zAxBCCOHWcoDHgbMBP0efXCl1FvAP4DJgBRDj6GsIIYSLkX5VCNFlyAgLIYQQTqO1/kRr/RlQdOQ+pdS5Sqm1SqlipdQypdSgNlziUeBvWutftdZ2rXW21jq7vXELIYSrkn5VCNGVSMJCCCFEh1NKDQX+C9wIhAOvAl8opXxO4BwewAggUimVoZTKUkq9pJRy+DeOQgjh6qRfFUK4I0lYCCGEMMNM4FWt9XKtdb3W+i2gGhh9AueIBryAi4FxwBBgKPCQg2MVQojOQPpVIYTbkYSFEEIIMyQAf2oYtlyslCoG4oFYAKVUulJKH+O2tOEclQ33/9Ja79daFwLPA5M6ujFCCOECpF8VQrgdty26qZS6AnhRax1pdixCCCGOsg94Qmv9RHM7tdZpLZ1Aa31QKZUF6EM3OyY8IYTodKRfFUK4HbccYdEw/+4SjI5bCCGESZRSnkopX8AD8FBK+SqlPIHXgZuUUqOUIUApNVkpFXiCl5gN3K6UilJKhQJ3AQsc2wohhHAd0q8KIboSt0xYAFcAHwF2swMRQgh3oZS6XCm1RSlVrpTaqZQa14qHPYQxxPh+YHrDzw9prVcBNwAvAQeBDGBGG8J6DFgJbAe2AGuAZr9dFEIIVyP9qhBCHJ/S2rxRXkqp2zA60oHA+1rrGYfsCwPeBCYAhcBftNbvteKcHsCnwFRghdZ6hMMDF0KILkYpdRbwBnAZsAKIAZCl7oQQom2kXxVCiJaZXcMiB3gcOBs4crmkl4EajGrFQ4AvlVLrtNablFLdgA+aOd/lDeeap7W2K6WcFrgQQnQxjwJ/01r/2vC7vKEWQoj2kX5VCCFaYOqUEK31J1rrz4CiQ7crpQKAi4CHtdY2rfVS4AvgDw2Py9VapzVzywX6A1cppRYBvZVSL3Zoo4QQws00jFwbAUQqpTKUUllKqZeUUkcmmoUQQrSC9KtCCNE6Zo+wOJY+QJ3Wevsh29YB41t6oNb6vsaflVKrtNZ3HOtYpdRMjDWr8fPzGx4fH9/2iDuI3W7HYnHX0iMGaaN7kDY61vbt2wtNXPUoGvACLgbGAbXA5xjzqB889EDpV12TtNE9SBsdqzP0q4f2qb6+vsN79OjR8ZE6mDv8O3aHNoC0w5W4QxvASf2q1tr0G8a0kDmH/D4OyD3imBuAdGfFMHz4cN0ZfP/992aH4HTSRvcgbXQsYJU2r48OxVjW7upDtl0ErDne46RfdR3SRvcgbXSsztav9unTx3lPRgdyh3/H7tAGraUdrsQd2qC1c/pVV03j2ICgI7YFAWUmxCKEEF2a1vogkIXx5rpps0nhCCFEpyf9qhBCtI6rJiy2A55Kqd6HbBsMbDIpHiGE6OpmA7crpaKUUqHAXcACk2MSQojOTPpVIYRogak1LJRSng0xeAAeSilfjNoV5UqpT4C/KaWux1gl5HxgjGnBCiFE1/YYEIGRUK4C5gFPmBqREEJ0btKvCiFEC8weYfEQUAncD0xv+Pmhhn23YCx1mg+8D9ystZYRFkIIYQKtda3W+hatdYjWupvW+g6tdZXZcQkhRGcl/aoQQrTM1BEWWutZwKxj7DsATO3AcIQQQgjRCrW1tWRlZVFV5TqfrYKDg9myZYvZYTiVtPHEeHh4EBISQkREhFtU3xfuzYx+1V36FHdoR2dpgxn9qqsuayqEEEIIF5WVlUVgYCCJiYkopcwOB4CysjICAwPNDsOppI2tp7WmtraWvLw8srKycIflQIV7M6NfdZc+xR3a0RnaYFa/KulmIYQQQpyQqqoqwsPDXSZZIcSRlFJ4e3vTvXt3ysvLzQ5HiBZJvypcnVn9qiQshBBCCHHC5E216AxkKojoTKRfFZ1BR/er0osLIYQQQgghhBDC5UjCQgghhBCiDaxWK7t27Trm/ptuuonHHnusXddIT08nLi6u1cenpaXxxhtvADB37lwmTJjQrusfKjU1lfT0dABmzZrF9OnTHXbuJ598kuuvv95h5xNCdE7Sr0q/eiQpuimEEEIIt5GYmEheXh4eHh5N22bMmMFLL73k8GvZbLbj7n/llVccfs0TMW3aNKZNm9bicTNmzCAuLo7HH3/8uMdt2uSY1eXT09OZPn06WVlZTdseeOABwCg8J4RwLdKv/s5Z/Wp7+77j9audnSQshBBCCOFW5s+fz5lnnmlqDPX19Ye9ue/M6urq8PSUt4xCdGXSrzqW9KutJ1NChBBCCNElzJkzh7Fjx3LXXXcREhJCUlISy5YtY86cOcTHxxMVFcVbb73VdPyMGTO46aabOOusswgMDGT8+PFkZmY27VdKkZGR0XTszTffzKRJkwgICOD7779nxowZPPTQQ03Hf/755wwZMoSgoCCSk5NZtGgRALNnz6Zfv34EBgaSlJTEq6++2uo2ffvtt/Tt25fg4GBuu+02tNaHtfeUU04BjOXo7rrrLqKioggKCmLgwIFs3LiR1157jblz5/L0009jtVqZMmUKYHyj+o9//INBgwYREBBAXV0diYmJLF68uOn8VVVVXHbZZQQGBjJs2DDWrVvX7HPT+Pw89NBDlJeXM3HiRHJycrBarVitVnJyco4aCv3FF1+QmppKSEgIaWlpbNmypWlfYmIizz77LIMGDSI4OJjLLruMqqqqVj9nQgjHcYV+dezYsdKvunG/KmkdIYQQQrTLo/M3sTmn1KnX6B8bxCNTUtt9nuXLl3P99ddTVFTEI488wuWXX86UKVPIyMjghx9+4KKLLuKiiy7CarUCxnzlL7/8klGjRnHvvfcybdo0li5d2uy533vvPRYuXMiCBQuoqanh3Xffbdq3YsUKrrrqKj7++GPOOOMM9u/f3zQEOCoqigULFpCUlMSPP/7IxIkTOemkkxg2bNhx21JYWMiFF17I7NmzOf/883nppZd45ZVX+MMf/nDUsd988w0//vgj27dvJzg4mK1btxISEsLMmTNZtmxZs0OX33//fb788ksiIiKa/Sbw888/5/333+fdd9/lhRdeYOrUqWzfvh0vL69jxhwQEMBXX3111NDlQ23fvp0rrriCzz77jLS0NP7v//6PKVOmsHnzZry9vQGYN28eixYtwtfXl7FjxzJnzhxuuumm4z5fQnQmHdGv9o7w4/GLhrT7PGb3q2+//TZTpkyRftVN+1UZYSGEEEIItzJ16lRCQkKabq+//nrTvp49e3LNNdfg4eHBZZddxr59+/jrX/+Kj48PEyZMwNvb+7BvsCZPnsypp56Kj48PTzzxBL/88gv79u1r9rrnn38+Y8eOxWKx4Ovre9i+N998k2uvvZazzjoLi8VC9+7d6du3b9M1kpOTUUoxfvx4JkyYwE8//dRiOxcuXEhqaioXX3wxXl5e3HnnnXTr1q3ZY728vCgrK2Pr1q1orenXrx8xMTHHPf8dd9xBfHw8fn5+ze4fPnx407Xvvvtuqqqq+PXXX1uMuyUffvghkydP5qyzzsLLy4t77rmHyspKli1bdlhssbGxhIWFMWXKFNauXdvu6wohjs2V+9XTTz9d+tUWdOZ+VUZYCCGEEKJdHDHywZE+++yzY861jo6Obvq58Q3jkdsOLfoWHx/f9LPVaiUsLIycnJzDtjd37JH27dvHpEmTmt331Vdf8eijj7J9+3bsdjsVFRUMHDjwmOdqdGQcSqljxnD66adz2223ceutt5KZmcmFF17Is88+S1BQ0DHPf7z2HLnfYrEQFxdHTk5Oi3G3JCcnh4SEhMPOHR8fT3Z2dtO2Qz9A+Pv7O+S6QriSjuhXT6TQo/SrR5N+tWPICAshhBBCiGM49Fs/m83GgQMHiI2NbfZYpdQxzxMfH8/OnTuP2l5dXc1FF13EPffcQ15eHsXFxUyaNOmwOdPHEhMTc1h8WutjfksJxrdnq1evZvPmzWzfvp1nnnnmuHEfrz1w+HNjt9vJyspqem78/f2pqKho2p+bm9vq88bGxh42p72xXd27dz/u44QQnYP0q8cm/erRJGEhhBBCCHEMCxcuZOnSpdTU1PDwww8zevToFr8ha851113H7NmzWbJkCXa7nezsbLZu3UpNTQ3V1dVERkbi6enJV199xTfffNOqc06ePJlNmzbxySefUFdXx4svvnjYG9hDrVy5kuXLl1NbW0tAQAC+vr5YLMbbwOjoaHbt2nXCbVq9enXTtf/5z3/i4+PD6NGjARgyZAjvvfce9fX1LFq0iB9++KHpcdHR0RQVFVFSUtLseS+99FK+/PJLlixZQm1tLc899xw+Pj6MGTPmhGMUQrgeR/er6enp0q+6cb8qCQshhBBCuJUpU6Y0VUq3Wq1ccMEFbT7XlVdeyaOPPkpYWBirV68+rODbiRg5ciSzZ8/mrrvuIjg4uKkyfmBgIC+++CKXXnopoaGhvPfee5x33nmtOmdERAQfffQR999/P+Hh4ezYsYOxY8c2e2xpaSk33HADoaGhJCQkEB4ezp///GfAeNO/efNmQkJCmDp1aqvbdP755/Phhx8SGhrKO++8wyeffNJUGO6FF15g/vz5hISEMHfu3MPO27dvX6644gqSkpIICQk5athxSkoK7777LrfffjsRERHMnz+f+fPnNxWGE0J0PFfuV//yl79Iv+rG/apqzdCYrmDEiBF61apVZofRovT0dNLS0swOw6mkje5B2uhYSqnVWusRHXIxB5F+1XU4uo1btmyhX79+DjufI5SVlREYGOjQc86YMaPZKu9mcUYbXY0z2nisf6+drV9NSUnR27ZtMzuMdnOHPtcZbTCjXzWjT3FGv+oOfWNna0NH9qsywkIIIYQQQgghhBAuRxIWQgghhBBCCCGEcDmyrKkQQgghRDPmzJljdghCCOFWpF8VJ0pGWAghhBBCCCGEEMLlSMJCCCGEEEIIIYQQLkcSFkIIIYSbs8uKYEKIVtBaY7dLfyGEcB1Sw0IIIYRwcw//XMmc3SsYHBfMJ79lc8/ZfbhgaJzZYQkhXMzfv9rKd1vzeXBSPx5bsJmEcH/uPacv/WKCzA5NCNFFyQgLIYQQwo1prcm2aX7cXsC/vsug3q6568N1rNh9wOzQhBAuZltuGRn5Nq6Zs5Ky6jp+21vMpa/+QlVtvdmhCSG6KElYCCGEEG6svMb4oJEY7s/9E/vy1R/HAbBm70EzwxJOMmfOHE455ZRj7v/000+Jj4/HarWyZs0aUlNTSU9P77gAhUuzVdfRMyKAK0bG88VtY/nHRYMoq6pjU06p2aEJYRrpV80lCQshhBDCjdmq6gCYeWoyN41PJjTAm25BvmzLLTM5MudITExk8eLFZofhsu655x5eeuklbDYbQ4cOZdOmTaSlpQEwa9Yspk+fbm6AwlS2qjpSogP5+4WDiAn2Y1iPEEASnF2d9KvHJ/2qc0nCQgghRIuUUulKqSqllK3hts3smETr2KprAbD6/l62KqVbIFvcNGHR2dTV1XXo9TIzM0lNTe3Qa4rmuWK/aquuO6yviArypXuIH2v2FpsXlBAnSPpV9yIJC+FSqmrreeG3KoY/9i2jnlzMoFlfM+rJxdz5wRpm/7yb/y7dzRs/7WJ15kGpYi1Ex7tNa21tuKWYHYxonbKGERaBPr9/COkbE8jOfBu19Xazwupw1dXV3HnnncTGxhIbG8udd95JdXV10/6nn36amJgYYmNjeeONN1BKkZGR0ey50tPTiYuL47nnniMqKoqYmBhmz57dtL+kpISrrrqKyMhIEhISePzxx7Hbjed6zpw5jB07lrvuuovw8HBmzZrFjBkzuOWWW5g4cSJWq5WxY8eSm5vLnXfeSWhoKH379mXNmjVN53/qqadITk4mMDCQ/v378+mnn7aq/Varlfr6egYPHkxycjLw+zenixYt4sknn+TDDz/EarUyePDgNj3P4oS5VL9aWlWL1efwmvzDEkL5TUZYiGa4Sr86d+5c6VfdmKwSIlyD1uiKIv758Q/4Fu7kxuQglIcnnl6+ZNWH8NnWbD5bm3PYQ3qE+XPP2SmcNzjWpKCFEML12aqNhMWh35r26xZETb2d3YXl9IkObP9Fvrofcje0/zzH020gTHyqzQ9/4okn+PXXX1m7di1KKc4//3wef/xxHnvsMRYtWsTzzz/PkiVL6NmzJzNnzmzxfLm5uZSUlJCdnc23337LxRdfzNSpUwkNDeX222+npKSEXbt2UVRUxIQJE4iJieG6664DYPny5Vx++eXk5eVRW1vLzTffzLx58/j6669JTU1l0qRJnHzyyTz66KM899xzPPLII9x999188cUXACQnJ/PTTz/RrVs3PvroI6ZPn05GRgYxMTHHjNfHxwebzYZSinXr1tGrV6/D9p9zzjk88MADZGRk8O6777b5eRadl9YaW3Udgb6HfzwYGh/C/HU57C+pJCbYz6TouqAO6Fd9wlPgvOfb/Hh36Fe///57QPpVVyYJC2EOux1y1sCObyBrBeRuQJUXcD+AN7Dv8MMfBuq79aQu9iRqYkewzGMEL6+u5I7317Blfyn3np2CUqrj2yFE1/J3pdRTwDbgQa11+pEHKKVmAjMBoqOjO0XRKZvN1inibKuVuUbCYuuGtZTvMQZWlpUZ30p9smQ5FgvsLrFzaR+vVvejwcHBlJX9PqXEp7YGS71zh+Daa2uoLjv2NJb6+nrKysrQWlNRUXFYfADvvPMOzzzzDH5+xgeuP//5z9x5553ce++9zJ07l2nTptGjRw/q6+u55557mDt3Ljab7ajzAFRUVODl5cVdd91FVVUV48aNIyAggN9++43hw4fzwQcf8PPPPwMQHh7Orbfeypw5c7j00kupqqoiJiaGGTNmUFlZCUBtbS3nnnsuffr0oba2lokTJ/LGG29wwQUXUFFRwbnnnstLL73U1MZzzjkHgPLyciZNmkRycjI//PADkydPpqqqqum4Yzm0XYc+X9XV1dTW1h73sc7WUuxtUVVV5cr/x4/brx7ap0ZGRjq1HVV1Gq0hP3sv6en7m7brYqNw7zsLf2ZIlAdf7a7FywKTkrzbdB136HOd0QYz+lVt1636/+bq/ardbm9zv9oYi9n9qjP6PmfqyH5VEhaiYxVsg9VvwYZ5UF4AykJVeD/WeY3gm9pwEhJ7MTDCg6EnjQF7PdRWQGkOFO/FI28jHpnp+Gyex9nAWd1HsCB5HH9Jr6Su3s79E/vhYZGkhRBOch+wGagBLgfmK6WGaK13HnqQ1vo14DWAESNG6MaiU64sPT2dzhBnW+Wv2gdr13PaKaOJC/UHoKbOztOrv2XutnrKGkZg/PnCMSRFWlt1zi1bthAYeMjIjHZ8Q3cijvfxqKysjMDAQJRS+Pv7Hx4fxjd3/fr1a9rer18/9u/fT2BgIAUFBZx88smH7QOwWq0cPHiQ/v37N53HZrPh7+9PeHg4oaGhTdsDAgLQWje9Oe3fvz8BAQEApKSkkJubS2BgIL6+vvTo0eOw+Ly8vIiLi2vaFhoaSkxMTNPvERER2Gw2PDw8CAwM5O233+b5559nz549TTGVl5c3nb/xuGOxWq1N+w99vnx8fPDy8jruY52t8e/oSL6+vgwdOtSh53SQFvvVQ/vUlJQUp/apeaVVsHgJg/unkDaqR9P2MXV2Xly3mDc21hIWoMgtrcXDorjnklOJCvQ94eu4Q5/rjDaY0a9WtvL/m6v3qxaLpc39auPvZverzuj7nKkj+1VJWIiOsfsn+PFp2P0jWLwoSTiThXUn8X5RL9ZneeLjaWHG2EQun5DCsqU/QtyI5s+jNRRuhy3zsWz+jPNy/48J/v68/8s4rtp+JXecP47hCUZH5+khJVqEcBSt9fJDfn1LKXUFMAn4l0khiVayNdWw8Gra5u1p4cOZJ/Py9xloNAs35LI0o7DVCYvOKDY29rDCaHv37iU21phSGBMTQ1ZWVtOx+/b9PsyvR48e2Gy2Vl8nIiICLy8vMjMzm96Q7927l+7duzcd054RgZmZmdxwww0sWbKEk08+GQ8PD4YMGYLW7a/rJCMVO5ar9auN9W6sR0wJ8fa08L+bx/De8r1s2V/Kbaf34qHPNvLpb9ncOD7ZjFCFi5B+tWXSr7affKITzpX9G8yeDG+diy7YzqbUP3Fd+NsM3vIHHstMpVu37sya0p+f7z+dv0zqh7dnC/8klYLIFDj1HrhpKVz/HT4DpnC11xLeLLmBNf/9I8Mf+oj+j3zNVf9dQVlVbce0U4iuRwPyKtwJNNawCPDxOGx7/9ggXp42jJevHEZcqB9LdxSaEZ5T1NbWUlVV1XSrq6vjiiuu4PHHH6egoIDCwkL+9re/NS01d+mllzJ79my2bNlCRUUFjz32WJuv7eHhwaWXXsqDDz5IWVkZmZmZPP/88w5b1q68vBylFJGRkQDMnj2bjRs3OuTc0dHR7Nmzp6mQnehwpvarje+ZAn2O/j6zV5SVv07pz/szRzN9dAIjEkL5cNU+h3ygE52D9KttI/1q+0nCQjhHxQGYfye8fjoUbmfniL8ywf4Ck1cPJ6PCjwcn9eOXv5zBa1eNYMbYnkRYfdp2nbjhqAtfw3L7KjwHXMCNngtYZr2fp3pv5eeMAv7yyQZ5MRWinZRSIUqps5VSvkopT6XUNOBUYJHZsYmW2arr8PY49qgzpRTjekfwy84i6txk1ZBJkybh5+fXdJs1axYPPfQQI0aMYNCgQQwcOJBhw4bx0EMPATBx4kTuuOMOTjvtNHr16sXo0aMBo6BaW/zrX/8iICCApKQkTjnlFK688kquvfZah7Stf//+/OlPf+Lkk08mOjqaDRs2MHbsWIec+5JLLgGM+eHDhg1zyDlF81yxX22uQO+xXDoinl0F5azOlNVDugrpV9tG+lUH0FrLTWuGDx+uO4Pvv//e7BBatv0brZ/upfWsUK2/ul9/t2aH7nn/An3as9/rxZtzdX29/bgPb1cbc9Zp/Wqa1o8E6cx/TtAn3feOfvfXPW0/n5N0ir9jO0kbHQtYpU3qH4FIYCVQBhQDvwJntfQ46Vddw/3/W6cH/fXL4x4zf122TrhvgV6792Crzrl582YHROZYpaWlDjvX5s2btcVi0bW1tQ47pyM4so2uyhltPNa/187Wr/bp08dxT0ozFq7P0Qn3LdBb9pe0eKytqlb3e/gr/ZdP1p/wddyhz3VGG8zoVzuyT3Fmv+oOfWNna0NH9qsywkI4Tm0VfHkPzL0YAiKov+F73g6+kZv/t4MB3YOZf9spnNEvGoszC2PGDILrF8PEp4kvW8di/wf4YcG7rNtX7LxrCuHmtNYFWuuTtNaBWusQrfVorfW3ZsclWqesqg4/j+MfM7B7MABbc0s7ICLX9Omnn1JdXc3Bgwe57777mDJlCp6eUupLOIcr9quNBXitzUwJOVKAjycnJ4Xzy84iZ4clOjHpV4UjSMJCOEZJNsyeCCtfh5Nvw3b1t1z/dTV//XwTIxLC+O+MkwhoxQugQ1g8YNSNqBt/wD+8O695PM2K1+9g8cacjrm+EEK4EFt1HX6ex08Ux4X64+tlYXte64uguZtXX32VqKgokpOT8fDw4D//+Y/ZIQnRocqaKdB7PCcnh7O7sJzckipnhiU6MelXhSNIiku0395f4cPpxgiLy+ZSknA2f/jvcjbllPLY1AFMH9XDnAq5kSl4zvyeyi/u4YYN7/Dj/7KpTfoEL//gjo9FCCFMYquqo6Up6R4WRXKklR35XTdhsWiRlGQRXVvjikJHFug9ltFJ4QD8uquIqUO7t3C06IqkXxWOICMsRPtsWQBvnQc+QXDDEooTJjDtzV/Zur+MV6cP5w+jE8xdzsfLF78L/8W2YQ8zxv4bla+eBWW55sUjhBAdrDUjLAD6RAeyI6+sAyISQrgiW3Ut/t4erV4Wvl9MEEG+njItRAjhVJKwEG236r8w7w/QbSBc9y0H/Hty5evL2Z5n49U/DOfM/tFmR2hQil7n/ok/ez+Md2km+r9nw8E9ZkclhBAdoqyqdQmL3tFW9pdUUdrK5aCN2lpCuDZZSrD1bNV1rapf0cjDohiVFM6vuyVh4SjSr4rOoKP7VUlYiLb55d+w4C7odSZc/QUHCOTK139lZ4GN168awWl9o8yO8DAeFsWA8RdwWdUDVJQeQL95NuRvNTssIYRwOlt1y1NCAPpEBQKwoxV1LHx9fSkqKpI318Jlaa2pqakhOzubgIAAs8PpFEqr6lq1pOmhRieFk1lUQU5xpZOi6jqkXxWuzqx+1S1rWCil0oCHMRIyL2qtPzU1IHfz6yvw9V+g3xS4eDZ25ckf31nBrsJy/nv1SZzSO8LsCJt17dhESisncsF3PszT/yDorfOwXPsVhCebHZoQQjiF1rphSkjLL/d9ohsTFmUMTwg97rFxcXFkZWVRUFDgkDgdoaqqCl9fX7PDcCpp44nx9PQkODiYiAjXfF/iamxVdQSeYIH0kxvqWPyys4iLhsc5I6wuw4x+1V36FHdoR2dpgxn9qtslLJRSfsCfgIla6xqz43E7K9+ARfdB33Ph4tloiyf/+GorP+0o5MkLBrpssgJAKcVdZ/UhNsSXyz618BFPEPjWFNQ1CyE00ezwhBDC4apq7dTbNX6teLXvHuqHRUHWwZa/KfXy8qJnz54OiNBx0tPTGTp0qNlhOJW0UTiTrbqOQN/WrRDSqG+3QHy9LGzZ33WXRHYUM/pVd/n/5g7tcIc2OIs7Tgk5GagE5iulPlVKdTM7ILex8RP48h7oMxEunk0dHtz14Vpe/XEXV47qwRUj482OsFUuO6kHfzjvbC6rvI+qijKjaKgt3+ywhBDC4coa6lG0poaFh0URFuBDoa3a2WEJIVyQrerEalgAWCyKhLAA9hSVOykqIURXZ2rCQil1m1JqlVKqWik154h9YQ0Jh3KlVKZS6spWnjYa6AVMAV4HZjk06K5q94/w6Y3QYzRcMpt6ixd3z1vHZ2tzuGdCH56YOsDc1UBO0LRRPRg0/BQuL7+HurJ8eO9SqJEXWyGEeymrNpYp9G1FwgIgwuotCQshuqiyqtoTrmEBkBjhz+5CeQ8lhHAOs0dY5ACPA/9tZt/LQA1GAmIa8B+lVCqAUqqbUiq9mVs3oBj4uWE6yBIgtSMa4tbyt8IH0yAsGa54n3oPX/780Tq+WJfDfef05bbTe3eqZAUY00MePT8Vug/njtrb0fvXwUfXQH2d2aEJIYTDlFUZfVprpoQARAb6UGCT2ZRCdEVlbRhhAZAYEcC+A5XU26VYpBDC8UxNWGitP9FafwYcth6SUioAuAh4WGtt01ovBb4A/tDwuFytdVozt1xgJdBPGZ+ghwC7OrBJ7qfiALx/GXj5wfSPwS+UF5bs4JM12dwzoQ83p3XegpW+Xh78Z/pwVniP5J/eM2HH1/DNg2aHJYQQDlNSWYvCzsiSRUZS9usHobzwmMdHWn0oLJMRFkJ0NfV2TVl1HcF+J1bDAiAxPICaerusFCKEcApXLbrZB6jTWm8/ZNs6YHxLD9RaFyqlPgV+ADRw7bGOVUrNBGYCREdHk56e3p6YO4TNZuuwOJW9jkHrHyG4JJs1Q5+gbE0G2w5s418rqhgb68kASzbp6dkOv25HthHghv4Wnl55KsmB+zhv+StsLfYmN+YMp16zo9toBmmjEOYrsdn4t9cLjM1eCbYeUJoN696H6xdDWNJRx0cEGjUstNadbuScEKLtGuvdtDVhAZBZVEF8mL9D4xJCCFdNWFiBI8sNlwCBrXmw1vpljCklLR33GvAawIgRI3RaWtqJRWmC9PR0OizOhX+G4o1wwWsMH3wZ32/L54XvfiMxIoBXbzylTcMGW6ND2wikAT7Ru7lr/uUM7VZA34xX6TtuKsQNd9o1O7qNZpA2CmG+hE2vMNhjJRsTZjBgxj8hfzPMmQwfTIfrvwXvw9dRj7B6U11nb9NqAUKIzquksu0Ji54RRj+yu6jcpVeLE0J0TmbXsDgWGxB0xLYgoMyEWLqmjf+DFa/B6Fth8GUs31XEjW+vpmdEAB/OHO20ZIVZrh6TyJQh8ZyXdx1VvpHw4XQoL2r5gUII4aryNjNg1+t8Wj+W/QlTQSmIToWL3oT8TfDLv496SITVB4BCqWMhRJdSWmnUuwlqQ8IiKtAHXy8Le6TwphDCCVw1YbEd8FRK9T5k22Bgk0nxdC2FO+CLOyB+FHWnP8JL3+3gurdWER/mx9zrRxEV5Gt2hA6nlOLJCwcSGRXDjMo/oiuKjFVR7HazQxNCiLb54SlqLX78g6vxshwyvaPXGZAyGZb9CyoPHvaQxoRFgdSxEKJLaRxhkZSzAN69GP47Eda+16pi5BaLIjE8gExZ2lQI4QRmL2vqqZTyBTwAD6WUr1LKU2tdDnwC/E0pFaCUGgucD7xjZrxdQm0lzLsKPH3g4tl8uCaXZ7/ZzuikMN65bhQh/t5mR+g0/t6e/Gf6cDbU9eA1v+sg41tY9qLZYQkhxIk7sAu2zGdZ6Plov/Cj95/+IFSXwvJXD9v8+wgLSVgI0ZWUVNbygOdckpfeDQd3Q0URfHYzfHAl1LU84ioxPECWNhVCOIXZIyweAiqB+4HpDT8/1LDvFsAPyAfeB27WWssIC2db/Kgxx/mC1yC4O5+vyaF3lJXXrxpBbIif2dE5XXKklecvG8JTRWNZFXAqesnfYN8Ks8MSQogT8+sroDz4yv88QvyaSTRHp0Ly6fDb22Cvb9ocEWgcKwkLIbqWwMyvmen5JeWDroZblsOty2HiM8YKap9cD/r4S5YmRPjL0qZCCKcwe1nTWVprdcRtVsO+A1rrqVrrAK11D631e2bG2iXs/B6W/wdGzoTeZ5JdXMmKPQc4b3Bsl6oWf3ZqN/56birXFF2FzSfKmBpSI98aCCE6iboa2DAP+k1hb23wsYvoDbvKWDVk5/dNm8L8vVEKWdpUiK6kppzhGx5ni70HauI/wMPTqHkzaiac+Shs/txIbh5HT1naVAjhJGaPsBCuovIgfHYLhPeGMx9ly/5SHvp0AwDnDYk1ObiON2NMIqP79+SW8uuNodWLHzU7JCGEaJ2d3xl9+qBLKamsPXYRvZRJ4B8Oa36fbenpYSHM35sCKbopRNex7gMCagp4tP4a/HyPqFM25g5IHAffPAQlx17KPrFhpZA9UsdCCOFgkrAQhoV/hvJ8uPA1Cqo9uPy1X1mx+wC3n96LhPCAlh/vZpRSPHnBQNZ5DGSRdSqseBV2/WB2WEII0bIN88AvFJLPoKSylhD/YyQsPL0h9QLY8Y1Rv6hBhNVHpoQI0VVoDSteI9svhR0+A44eUWuxwHkvQn0tLH7kmKdJbHivKCuFCCEcTRIWAjZ8DBs+gvH3UR8zlEe+2EhlTT2f33YKf5qQYnZ0pokM9OFPE1K4s/B8yq2J8PmtUFVidlhCCHFstVWw7SvoPxU8vSmprD32lBCAvudCbYUxKqNBuNWbA+UywkKILiFzGRRs5bugqQQfq7B6WBKMvsl4v5i3udlDooN88PG0sO+gTAkRQjiWJCy6uvIiY3RF9+GsTbyW815aysINufzxzN70irKaHZ3ppo3qQUJ0BHfX3oQuzYavHzA7JCGEOLY9S40ERMokauvtVNTUHz9hkXgK+AbDlgVNm0IDvDkoCQshuoYtX4CHDz94jiHweH3FmDvAJxC+f6LZ3Uop4sP82VtU4aRAhRBdlSQsurpvjKXt5ic+wAWvLKfQVs3LVw7jlrRksyNzCZ4eFh45rz9fl/RgVferYc27h30TKYQQLmXH1+DpBz3HUVJZC3DsKSEAHl7Q5xzY/lXTaiFh/t4cqJCEhRBuT2vY+iUkn05Bjdfxk5v+YTDmdti6AHI3NHtIjzB/9h6QhIUQwrEkYdGV7foB1r1P8dBbuOeHWk7pFcHiu8czeVBMl1oVpCVjkiOYPCiGa3afRnVQInz5J2PYtRBCuBKtYfvXkDQevPworjASFsf9EALQe4JRpDNnDWCMsCiprKWu3u7siIUQZtq/Dkr2Qb9zKW1p+hjAyBvAyx9++Xezu3uE+bPvQAW6hSVQhRDiREjCoquqrYIFd0FoT+7Nn4C3p4VnLxlMoG8LL1Zd1N/OSyUgIICHaq4xVg1Z+n9mhySEEIcr3AHFmUYCAppGWBxzlZBGSWnG/S5jedPwAG+0/v3xQgg31ThitPfZxopCvp7HP94vFIZMg40fQ1neUbvjQv0oq66TvkMI4VCSsOiqfnoWDuyk8LR/8G1GKdeO7Ul0kG/Lj+uiwq0+PH3xYD4q7s3ubhNh6fNQtNPssITocEqp3kqpKqXUu2bHIo6wu2Elo4YERGnjlJCWEhYBEdBtEOxMB4wRFoAU3hSiA5jap+5ZCpH90AERrRthATD6ZmPFkJVvHLWrR5g/gEwLEUI4lCQsuqL8rbD0nzDoMt4rSEJruHh4nNlRubxTe0cwPCGUOw5egvb0hS/vNoZgC9G1vAysNDsI0Yw9P0FQd6OiP7+PkGjVh5Dk02Dfcqi2EeYvCQshOpA5fWp9Lez9FRJPoaKmnjq7bl1fEZ5s1L1ZPRvqDu8jeoRLwkII4XiSsOhqtKbkk7sox5dnuZq3f8lkTHI48Q1ZcXFsSiluO70XG0p8+SHuJtiVDhv/Z3ZYQnQYpdTlQDGwxORQxJHsduPb0p6nQkMNosaEQ8ixlio8VFIa2I0PMKEBxoeWg1J4UwinMrVPzVkLteWQeErT//VWJSwARlwL5QVGsd5DxIdKwkII4XgtTFYT7kZvW0hw7jIeq5/B26tLSI60cs/ZKWaH1Wmk9Ylk6pBYrl1rZ03MAIK/fsD4psFHloAV7k0pFQT8DTgduP44x80EZgJER0eTnp7eIfG1h81m6xRxHk+AbQ8nVRSxtSqS3Ia2/Ly5Gj9PWLfiZ8rLy4/bRo+6ak7Bwt6fPmB7rPFdxq9rNuJbuK0DoncMd/g7tkTa6D7a0qdGRkY67LmJ2/c5vYBlWZp1W38B4MC+HaRX7Gr5wdqD0T4RVHz7POvzgw/bFegNKzbupD9Zx3y4O/yN3aENIO1wJe7QBmeRhEVXUldN9ZcPsM/enZ7n3MaOsb3NjqjTUUrx1EWD2F1UwY15l/GB5WH4+QU4/UGzQxPC2R4D3tRaZx1vFSGt9WvAawAjRozQaWlpHRNdO6Snp9MZ4jyu5a8B0Pec6+kbmgDA6xm/0iemntNOG9u6Nu4aQoLKIfqM8dyVvoiI7omkpXWe1wm3+Du2QNroVk64T01JSXFcn/rx2xAUx5izLyBzxV5YuYHzTj+ZuNBWjrhVN+Cb/hRpg3tCQ58DkLzpZ+p8PElLG3XMh7rD39gd2gDSDlfiDm1wFpkS4mYKyqoZ/eQSVu45cPTO5a/iW7aH5y1Xc+FJPTs+ODfh6+XB638Yzm7f/iz2GIde9i8oOfY3CUJ0dkqpIcCZgCyP46qyVoK1G4T0aNq0M7+c5MiA1p8jYQxkrcJX1RHg7cGBcqn0L4QzuESfmv0bdB8KwJ7Ccrw9LcQG+7X+8UOnG9PP1rxz2OYeYf4yJUQI4VCSsHAz67OKyS2t4uNVxgfo1ZkHuXXub9iKcrD/8DTf1Q+h5+jz8feWwTXtERXky8tXDuPRioupq6+HJY+ZHZIQzpQGJAJ7lVK5wD3ARUqp38wMShwiexXEjWiqX2GrriO3tIrkyBOYrtZjNNRXQ84aQgO8pYaFEM6Thpl9auVBOLgbYocBsKeonB5h/lgsxx7pcZTgOEg6DdZ/aNTQaRAf5kdOcSV19fbjPFgIIVpPEhZuJiPfBsCSrXnU2zVv/7KHLzfsZ+3bf0bXVPAP+3SuHpNobpBuYkRiGOeljeb12nNg/QfGtxVCuKfXgGRgSMPtFeBL4GzzQhJNyovgwC4jYdFgV4HxWnBiCYuTjfu9vxAe4C2rhAjhPOb2qTlrjPvYxhEWFSSGn8BorEaDLoPivcYKQw16hPlTZ9fsL6lyRKRCCCEJC3fTmLAotNWwYvcBvt+az6iA/Zxc/CVv1Z3FgMEjiQ7yNTlK9/HHM/rwfeR0igimZuFfZJlT4Za01hVa69zGG2ADqrTWBWbHJjBGVwDEndS0aWdDwqJX1Al8CAmIgNBEyF5NqCQshHAa0/vU3I3Gfcxg7HZN5oFyEsPbsFpc38ng5W+MsmjQuOrcPpkWIoRwEElYuJmMAhupsUF4eSj++vlGSqvqeDHqS7S3lcSLHmXWef3NDtGteHta+PsVY/hn/SV4Z/+K3vKF2SEJ4XRa61la6+lmxyEaZK0EZWn6thSM+hUeFkWPsBP81jR2GGSvIcxfEhZCdJQO71Pztxg1b/zDyCuroqrWTmJEG0ZY+FiNpMWmT6HO6C96hMnSpkIIx5KEhRvRWpORb2NojxD+NCGFHfk2RnlmEL3/OzzH/ZHTh/Yl0LeVa2yLVusVFUjvs29mu707pV8+AvZ6s0MSQnQlWasgKhW8f//AsS2vjIQwf7w9T/BlvvswKM2ih4+NQls1WkaNCeF+8jdBtPEF1q6CcgB6tiVhAca0kKpiyPgWgJhgP7w9LU0jfoUQor0kYeFGCsqqKauqo1eklZvGJ/Pk1AH8M/ILCIiEUTebHZ5b+8PYZBZFXUdw+W7yf37L7HCEEF2F3Q7Zqw+rX2G3a1buOcCwhNATP1/34QAMUjuprrOTX1btqEiFEK7AXg8F2yDKSFis3VcMwIDY4LadL+k08I9omhbiYVEMiQtpfrU6IYRoA0lYuIEiWzVVtfVN2exeUYEAXBmRQczBVXDqn41he8JplFJc9odb2EQS9u+forZGik0JITpA4XaoLj2sfsX2/DKKK2oZ1TPsxM8XMxiUheTa7YCx3KEQwo0c2A11VYclLJIiAgj2b+MIXA9PGHgxbFsEVSUAjE4KY2NOKWVVsjSyEKL9JGHhBi76zzKuf2sV32zOA6BPN6tR/HHJ3yC4BwyfYW6AXUR0sB+Vp/yFbvY8fvjgebPDEUJ0BU0FN38fYbF8l/HN5uik8BM/n3cARPYlqmwzAJlFMg9dCLdSsNW4j+yL1pq1+4oZHB/SvnMOutRYEnmzUcdrVFI49XbNqsyD7TuvEEIgCYtOr6Csmj1FFSzNKGTOsj1cOaoHUYG+sPlz2L8W0u4HTx+zw+wyRpxxCbv9BzJg52ts3ptndjhCCHeXtRJ8giG8d9Om5buLiA32JS7Ur23njB2Gb8E6PC2wp0hGWAjhVop2GPcRvckpqaKgrJoh7U1YxA6DsOSmaSHDeoTi5aGakqdCCNEekrDo5DblGMPvYoJ9iQ7y4b6z+xrzE79/AiJSYPDlJkfYxShF5HmP000d5NcPn8Zul4J1QggnylplFMq0GC/n1XX1/LKziNFJ4Sil2nbO7kNRFUWcFGKTERZCuJvCHWCNBt8gVu42EgrtTlgoZRTf3LMUSrLw8/ZgcFwIy3cXtT9eIUSXJwmLTm5TTikAX94xjm/vHm/MQdz8mTGvOe1+sHiYG2AXZO2bRl7kGM63fci7P24yOxwhhLuqrTKWJ+w+rGnTVxtyOVhRy9Sh3dt+3ljjfKf472W31LAQwr0U7mgakbVg/X6ig3wY2L2NBTcPNegSQMOGjwEYlRTG+qwSyqvr2n9uIUSXJgmLTm5jdgmJ4f6EBXgT5OtlVIz/8VmI6AP9zzc7vC4r6vzHCFdl5C55mW25ZWaHI4RwR/mbQddDt4FNm976ZQ89IwI4pVdE288bPQA8vBli2U1mUbksbSqEOynaARG9KKmo5Yft+Zw7KBaLpY2jsQ4VlmSsMrTpEwBG9TTqWKyWOhZCiHaShIUT1NbbeXzBZvLLnL9SxMacElIPzYxvW2i8iR13j4yuMJGKG0FNYhrXenzJ8wvXmR2OEMId5W4w7rsNAmB3YTlr9hYzbVSP9n0A8fSG6AEk1W6jvKaeQluNA4IVQpiuvAgqD0JEH9K351Nbrzl3UIzjzj/gIti/DgozGJ4QiqdFybQQIUS7ScLCCTZml/DG0t18s6n9RRdfXLKDiS/81PQN1/+truK5b7YBUFxRw74Dlb+vna01/Pg0hPY0XjSEqbzT7iWCEqJ3zmPL/lKzwxFCuJvc9eAdaPT5wJq9xjeZ43pHtv/c3YcRWbYFhZ2dBbb2n08IYb6Du437sCS25ZbhaVGkxjpgOkij1AsABZs+IcDHkyHxIaRvK3Dc+YUQXZIkLJwgu7jysPu2KrRV85/0nWzZX0pOSRW5JVWsK6jn111GtrqxfsWA7kHGA3Z8a2S2x/3JWBdbmCtxLHVxo7nZcwGvp281OxohhLvJ3QDdBjQV3Fy3rxh/bw96RVnbf+5ug/CsKydOFci0NiHcxcE9xn1IAjvybSRGBODt6cCPAkGxkDDGqGOhNWendmNTTil7pXivEKIdJGHhBNkHjURFTjsTFi99l0FlbT1gvBH9bms+AFkN59+YbawQkhob/PvoiuAesjKIC/Ec/2diVBE+mz7mQLkMqxZCOIi9HnI3Nk0HAViXVcLA7sF4OGI+ercBAJzkk81WSVgI4R6KM437kB5k5Nvo7Yjk5pEGXAiF2yB/M+cM6AbAok37HX8dIUSXIQkLJ2hMKDQmLo7Hbtec/lw6H67ce9j2l7/PYM6yPVw0LA4vD8W6rN8TFrmlVdTU2dmYU0r3ED/CArxh9w+QtRJOuRM8vBzeJtFGvc6gMnIgN1g+57Pf9rZ8vBBCtMaBXVBbDjFGwqKmzs7mnFIGt3d5wkZR/UFZGGPdz7ZcmdImhFs4mAn+EVRZ/MgsKndOwqLf+aA8YOP/iA/zZ0D3IBZuyHX8dYQQXYYkLJzgRKaE7C+tYldBOV8fUu+ivLqOZ7/Zxtmp0fzjooH0iwliWUYRP2cUEuhlDKbIKa5kU3YJqbEN00F+eg4CY2DINKe0SbSRUviddi9Jllz2L3tfqu0LIRwjd71x37BCyNbcUmrq7QyOC3HM+b38ILw3Azwy2Z5nk75LCHdQnAmhCewuLMeuoXd0oOOvYY2EpPGw8X+gNRMHxLB2XzH7S9o36lgI0XVJwsIJGkdW5JVWUVtvP+6xmUXGGve/7T2I3W68ITxQXoPWcEbfaDw9LAyKC2ZDdgm19XamJHsDxpvTXYXlxtrZOWth948w+mbw8nVew0Tb9D2XEmsyF5Z/SPq2fLOjEUK4g/3rweIFkf0AWLO3GIDB8Q4soNdtIHE1u7BV1/H019v4bmv7C0kLIUx0MBNCEtieZ0zz6h3thBEWYBR+P7gHcn5jYuO0kI25bMgq4aHPNrCv7PjvjYUQ4lCSsHAwrTXZxZVYfTyxa8gtOf7SppkNhYiKK2rZVWhr+hkgxN+Y2tH4jdkdZ/RmWLSxVOk3m403jgO6B8MvLxmV4ofPcHRzhCNYLAScfg/9LPtY9vU8s6MRQriD3PUQ1ddYghT4OaOQuFA/4kL9HXeNbgOwVuYQhI3/pO/kT/PWUV5d57jzCyE6jr0eSrIgNIGN2SV4e1pIjnRSwqLvuUZCdeMnJEVaSYkO5NmvtzHlpaW8++tenlpR2VSHTQghWiIJCwcrrazDVl3H8IRQoOVpIZmHVE5enWksSXewwijOGOJvvBGdMjiWf142hFvSkgn1UXhYFF9tyMWiYEhQGWz8BIZfDb4O/GZNOJTnoIsp94lkXOGH/N+327HL8GohRFtpbYywaCi4WVdv55ddRYxNjnDsdaKN6SanhxZw3uBYDlbU8vYvmY69hhCiY5TmgL0WQhLYkF1Cv5ggvDyc9DHALwR6n2W8P7XbuWJkPH7entwzoQ9f3nEKvh6KK1//lfVZxc65vhDCrUjCwsGyio0ExMieYUDLK4VkFpWTFBFAqL8XK3YbCYviSmOERWjDCAtfLw+mDu2Op4cFD4siJtiXytp6xvWOJHT9m6CUMR1EuC5Pb3zG3MypHhtY9N0S5u+sNTsiIURnVZYLFYVNCYuNOaWUVdUxtreDExYN9TH+Od6DF68YyrjeEbz7qyQshOiUGlYIsQcnsCm7lAGNNdCcZcBFUJYD+35lxtierHroTG47vTepscHcP9KXID8vbn9/jXNjEEK4BUlYOFhj/YqTEsMO+/1YMosqSAj358x+0Xy+Npv1WcUUHzHC4khxoX4AXDogEH57C1IvhOA4RzVBOInnSdegvfz5a/j3LNpTS0mlJC2EEG2Qu8G4b0go/JxRCMCY5HDHXicwGgIijeVTMaYg5pVWSQFOITqjg0bCIkdFUVZdZ9RAc6Y+54Cnn1F88wiR/hamjUogs6iC0ip5LySEOD5JWLRTaVUt57+0lNWZBwBjyVGAxAh/Qv29mn5vjtaazKJyEsIDeGhyfyIDfbh73joOlhudd7Bf88uT9oywEujjyYTKRVBjgzG3ObhVwin8w1BDp3NyxXdY6w7yyOcbm4quCiFEq+UZCQSiUwFYsfsAvaOsRFh9HH+t6AGQZyRIwvy9qbNrbFLHQojOpzgTUGyqMEZW9Itx8ggLHyuknAObPoP6o/uMXg1Lqmbk25wbhxCi05OERTt9vzWfdVklzF+3H4BCWw1KGW/sAnw8qaipB6Cmzs7z32xr+lZ9Y3YJf/lkA+U19SSE+xPs78WMMYlk5NvYe6ACq48n3p7N/3numdCHT24cgdeqV6HneIgZ3DGNFe036iYs9joeCVvC5+tyOPufP7JsZ6HZUQkhOpO8TRAUB34h2O2aNXsPNtVNcrhuAyF/C9TXNhWCbiwMLYToRA7ugaDuZBQZo3gbEwZONeAiY/ranh+P2iUJCyFEa0nCop0WbzGWqVzVMMKiyFZNqL83nh4W/L09qGxIWKzOPMiL32WwaON+6u2aez5axwcr9wGQ1FClubG6+6ackqY3hs0Jt/rQu+BbKNsPY253WtuEE4QnQ9/JnF3zLT/fPZoeYf5cN2cVW3NLzY5MCNFZ5G+G6P4A7Cq0UVpVx7AeTkxY1NdA4Q5CG6YpHiivcc61hBDOU7wXQhPYmW8jJtiXAB9P51+z11nGKnbNTAuJD/XD28PCTklYCCFacNyEhVLqw0N+vsb54XQuNXV20rfl42lRbM4ppayqliJbDRFW402dn7cnFbVGwmLfAaMY55b9ZXy0ah9bc8v4+4UDef2qEYzrZRRKiw3xBYxs8/ESFgAsfxUi+kCvM53UOuE0J9+GV10ZsXs+5d3rRhHg48GdH6ylquHfihCOIn24G6qrgcLtEGUkLH7bWwzAsIQQ51wveoBxn7uB0ADjdalxJSshuqJO26+WZENwHDsLbB0zugLAyxf6nQtb5kNd9WG7PD0s9IwIkBEWQogWtTTC4myllGr4+QVnB+MISimLUmqOUuonpdRSpVRfZ11r1Z4DlFXVMW1UD+zaeONYVF5NeIAxj9jfy4PKGmPe3r6DjQmLUt75NZOB3YO5/KR4zuofjcViPMXdG4pp1tl10zdZzcpaDTm/wciZxgohonPpMZrSwN7w6ytEBfrw9MWD2JpbxnPfbDM7MuF+Ol0fLlpQtAPsdU2JhDV7DxLk60lShJM+gET0AQ9vyNvY9LokU0JEF9f5+lW7Hcpy0IGx7CwoJzmygxIWYEwLqSqBnd8dtatXtJWMAklYCCGOr6XxYD8BvyiltgO+Sqm3mztIa32VwyNruyGAj9Z6nFJqHHA3MNMZF1qfXQLATWnJvLt8Lyt3H6DIVkP/hqWi/L09yC013tjtbRhhsSG7hIqaev50Vh/UEcmGiAAfvD0s1NTbj1lwE4AVrxpD7AZf7oRWCadTiuzukwna+k/Ylc7pfU9j+ugevP7Tbk5LiWJMLwcvTSi6ss7Yh4vjydts3DdMCVmzt5ghPUKbEt8O5+EJkSmQt6kpYSEjLEQX1/n61YpCsNdR5h2JrbqO5MiAjrt2Uhr4hRrTQlImHrarV6SVhRv2U1Vbj6+XR8fFJIToVFoaYXEJ8BKwDdDAzmPcXEkWoBqy36GA0yoabs8rIzrIh5hgP3pFWtmyv5RCW3VTpXa/Q2pYNE4JaSzCeUrvoz+UWiyKmIZpIccaYeFVUwybPoUhV4JPoKObJDpIQeRY8A+HlW8A8OCk/iRHBnDre7+xu1BWDhEO47A+XCn1rlJqv1KqVCm1XSl1vZNiFseTvwksnhDem5o6Oxn5NgbEOrnaf/QAyN9MkJ8XSsFBqWEhurbO16+WZgOQh7H0cY/wDkxYeHhB//Nh60KoqThsV68oK1rDrgJ53yOEOLbjjrDQWlcB7wIopby01o868uJKqduAGcBA4H2t9YxD9oUBbwITMJIOf9Fav9eK0xYCtcBWwBcY68iYD7Ujz0afaCNp0DMigE37SyitqiM8wEg2+Ht7NCUo9h2spFeUlYx8G4G+ngyKC2n2nLHBfmQWVRB6jBoWsTnfGAXQRt7g+AaJDmP38IZhV8HPL0DxPvxC4nnj6pO48N8/c/1bK1l056l4eUhNXNE+Du7D/w5cp7Wubphql66UWqO1Xu2IWEUr5W02pml4erNzfyl1dk1KNycnr6NTYd37eFQeINjPi4MyJUR0YZ2yXy3NASDHHgrU0b3hy7EOM+AiWD0HdnwNqRc0bW5aKaTA1jQ6WQghjtRS0c2kxhvw9qG/H7GvrXKAx4H/NrPvZaAGiAamAf9RSqU2xNVNKZXezK0bRoKjTmudAlwEPNeO+I7Jbtdk5NvoHWW8UUyMCGDfgUrAWMUDwN/bk8raeipr6ikoq+as/tEAjEkOx+MYw3djQ4w6FsHNjbCoryU2ZxEknw4RvR3dJNHRRlxr3K+eDRhJr39cNIidBeV8+lu2iYEJd+HIPlxrvUlr3Vg1TTfckp0VuziG/M1NBTe35ZYB0Lebk9/oN1yP/E2E+XvLlBDRpXXKfrUhYbGrJhj4/b1mh0kYC9boo1YL6RkRgEXJ0qZCiONrqYZFBkbnqRruGx35e5smnmmtPwFQSo0A4ppOrlQARrJhgNbaBixVSn0B/AG4X2udC6Q1d86GqSBFDb8WAsFtia0l2cWVVNbW0yfayA73jPBv2hfesEqIr5cxJSSroeBm326BPDS5Hyclhh3zvI2FN5sdYbF1AT41RUaxTdH5hfSAPufA6rdg/H3g6cNZ/aMZHBfMC0t2MGVwLH7eMqdTtItD+3Cl1L8xRsX5AWuAhQ6JUrROVQmU7IMRxsIEW3PL8PJQJDl7PnrjSiF5mwjxHyRFN0VX1/n61dJssHiys9yPUH8v/L07YEnTQ1k8jJEVq2ZD1e/LuPt6eRAf5i9LmwohjqulKSFNIzAalm46E5gFZAIJwF+BJU6Iqw/GKInth2xbB4xvxWO/BWYopX4AfDCKbjZLKTWThoKc0dHRpKentzrAtfnG6h+27B2kp++i+MDvS1Lu3b6J9IKt5GXXUFNv54v05QAU7NlKrxAPDu7cS/oxZjeW5TUU6czYSnppxmH7hqx5Bi/vSFbmeMP+1sfa2dhsthP6W3RGjW0M9RnF4IqFbPn47+R1SwPg7Jh6nllZxcUvfMMfh/ni7dE5V4LpSn9HV+XoPlxrfYtS6nbgZIykcfWRx7SnXzWLq/8dGwWVbGEYsCHPTlF6Oss2V9HNX/HzTz+2+Nh2tVFrxngFU7R2MfWVPdlXrF3y+eosf8f2kDaar6P71UP71MjIyDY9N313rCHEK4z1u/YT5GnO/9+gmp4Mq69my2fPYbOe1BRDiKWatbtzXfpv3hxX/3faWtIO1+EObXAarXWrbhjFLP2O2OYPZLX2HMc59+PAnEN+HwfkHnHMDUB6e691rNvw4cP1ifj39xk64b4FuqSyRmutdV5JpU64b4FOuG+B3l1g01pr/fqPO3XCfQv0S9/t0An3LdB5pZUtnnf9vmKd9Jcv9c78ssN35G7S+pEgnfHW7ScUZ2f0/fffmx2C0zW1sb5e6xeHaf36GYft/2jVPp1w3wL98GcbOj44B+lSf8cOAKzS7etnHdqHA68AdxzvmBPtV83Saf6trnhD60eCtD6YqbXW+uQnF+s/vv9bqx7a7jbOmaL1q2n67g/X6pOfXNy+czlJp/k7toO00bE6W7/ap0+ftjV09mSt35igz3wuXd/w1sq2naO97Hatnx+g9bsXH/Y3fnLhZt3rgS91bV29OXG1kbv8X5R2uA53aIPW7e9Xm7udSFU/C5B4xLYE2jgdpAU24MhJuUFAmROu1SY7C2x0C/IlyNeYuhEZ6ENAw/D9xikhjcP5sw421LYI8GnxvAPjgln/yASSjlwj+7e3wMOb3G5nOqoJwhVYLHDS9ZC1EnLWNG2+eHgc147tydu/ZPL1plwTAxRuxNF9uCdSw6Jj5W8GnyAIjqe0qpackipSnF2/olF0KuRvIdzfIkU3hfhd5+hXS3MgKJac4sqOr1/RSCkYcAHs/A7P2t+nhfSKtFJbr9l7oOI4DxZCdGUnkrD4P+A7pdSTSqmblVJ/xxjy9n9OiGs74KmUOrSy5GBgkxOu1Sb/uGgQn9/2+wIkSikSwgPw9rRg9TFm2vg3JCz2l1QS6Ot5zEKbRwrwOWKmTm0lrHsf+p1HrbdUUXY7g68AL39jbuch7j0nhZToQG58ZzV/+WSDScEJN9LmPlwpFaWUulwpZVVKeSilzgauwDlTAsWx5G2GqH6gFNkNifCEcP8WHuQg0alQV0kPlU9lbT1VtfUtP0YI9+f6/arWUJpDnbUb5TX1RAa2/OWZ0wy4COx1RBYsa9rUtFKI1LEQQhxDqxMWWutnMIoCRQNTgCjgWq310229uFLKUynli5GJ9lBK+SqlPLXW5cAnwN+UUgFKqbHA+cA7bb2Wo3lYFNFBhy8L1TvaSkywL0bdT/DzMhIPuSVVhBxjmdJW2fy5UWxt+Iy2n0O4Lr8QoxjVxv9B9e8v2L5eHnx261imjerB+yv2sjG7xLwYRafXzj5cAzdjDH8+CDwL3Km1/sI50YqjaA35m5pW7MgtrQI46nXIaaJTAUio3Q0ghTeFoJP0q5UHoa6SSl9jpbpgv3a8H22vboMgIoVuuelNm5IbEhY7JGEhhDiGVicslFLewFiMwpdpwKnA2IaEQ1s9BFQC9wPTG35+qGHfLRgVk/OB94GbtdYuM8KiOQ9M6sfrV41o+r1xhEVOcWX7XiBWz4GwZEg8pZ0RCpc17GqoscGmTw7b7OftwX0T+2L18eSNn3aZFJxwB+3pw7XWBVrr8VrrEK11kNZ6oNb6dedGLA5TmmMkrhsSB3klRsKiW3AHJSwi+4KyEFNt9EOFtqPqrQrR5XSKfrVhSdMyHxdIWCgFgy8juHQLHDCSn0G+XnQP8WNzTmkLDxZCdFUnMiXkP8DpwO3AiIb7NODfbb241nqW1lodcZvVsO+A1nqq1jpAa91Da/1eW6/TUaKDfOkTHdj0e2MNi9KqOkL8vNt20vytsPcXY3SF6pyrRYhWiB9pfCBY/dZRu4J8vbj8pHjmr9/P7sJyE4ITbsLhfbjoQPmbjfuGhEXjCItIawcN7/byg/BeRJTvAKCgTBIWQtAZ+tUyow5WiUc4QPtG/DrCwEvRKFg/r2nT0B4hrNl70MSghBCu7EQSFlOBc7XWX2mtN2utF2FM05jqjMDcgZ/X7zWXgtv6AvHbW2DxgiFXOigq4ZKUgmFXQfYqyDt6INHM8Un4eXnw6PxNjZXEhThRU5E+vPNq7Bei+hm/llYRYfXG2/NEXsbbKao/1pJtAOSXVXXcdYVwXVNx9X7VlgdAkQoFTB5hARAST3HIAFj/gTHVDRjaI5SckipyS6RfEUIc7UTe6eRiLNV0KD9gv+PCcS+NU0IAQtryAlFbBWvfg35TICDCgZEJlzTocvDwht/ePmpXVKAvd57Zm/RtBSzZkm9CcMINSB/emeVvhqDu4Gd86Mgtqeq4+hWNogfgWZJJAJXkl8oICyHoDP1qufGeoUAbRdtNT1gAedFpcGCXsUIaxggLgLX7ZJSFEOJoJ5KweAdYpJS6QSk1USk1E1gIvK2UOr3x5pwwOyd/799X+2jTC8SWL6CqWIptdhUB4UZyat0HRrLqCFePSaRXlJW/LdgsFfpFW0gf3pnl/V5wEyC3tJpuHZ6wMKajDPPLJV+mhAgBnaFftRWAlz8Hao2pyW2eouxABZFjwNPXeL8DpMYG4e1hYc3eYnMDE0K4JM+WD2lyY8P9A0dsv6nhBkbF46T2BuUu/A4dYdGWKSGr50BYEiSOc1xQwrUNu9pYLWTLFzDo0sN2eXlYePS8VKa9sZw3l+7m1tN6mRSk6KSkD++s6muhcDv0OqNpU15pVdO3kh2mIWEx3DebLTIlRAjoDP1qeT4ERFJSUYNSEOh7Im/9naPe0x/6nmu83zn7SXy8fOkfGyQJCyFEs1rda2mtezozEHd0+JSQE8xoH9gFmT/D6Q+DpQPnKAtzJY6D0J7GtJAjEhYAY3tFcFpKJP9dupvrTumJ7yF1UoQ4HunDO7GiDKivgSgjYVBdV8+B8pqOH2ER0gO8A+lv2ccPMsJCiM7Rr9rywRpFSWUtgT6eWCwuUsB9yJWw8WPYthAGXMjQHiG8v2IvtfV2vDzkfa8Q4nfSIziRl4cFLw/jhSHoRKeErPsAUDD4CscHJlyXxQLD/gB7foKinc0ecuP4ZIrKa/h4dVYHByeEMEVjwc1oY0pIY/2IDk9YKAXR/Um2Z0oNCyE6i/ICCDASFiH+5k8HaZKUBkFxsHYuYBTerKq1sy23zNy4hBAuRxIWTta4UsgJTQmx22Ht+5B8GgR3d1JkwmUNvhKUBda93+zuUT3DGBwfwis/7KS6TmpZCOH28jeDxRMi+gC/L2kaHdzBCQuA6FRia3ZTUFYlKxYJ0RnY8sEaSXFlrUsU3Gxi8YAhV0DGEijJZmh8CIAsbyqEOIokLJyssY7FCSUsMpdCyV4YMs1JUQmXFhQDSacZo2zs9qN2K6X401l9yDpYydvLMk0IUAjRofI2QXhv8PQBIKe4EjBhhAVAdCp+9WWE1xdSUlnb8dcXQrRefR1UFEFAFAcrattWT82ZhlwJaFj3PnGhfkRYfaSOhRDiKJKwcLLGlUJOKKu99j3wCYa+k50UlXB5Q66Ekn1G8qoZp/aJZHyfSJ79Zhs3vL2KnQW2Dg5QCNFh8jY3TQcB2JBVgo+nhZ4RAR0fS/QAAPpa9spKIUK4uooiQIM1iuyDlcQG+5kd0eHCkiDhFFjzLgpjedM1+4rNjkoI4WIkYeFkTVNCWlt0s7oMNn8OAy4ELxd7YREdp+9k8AkypgYdw1MXDWTqkO6s3HOAi/+zjI3ZJR0YoBCiQ1SVGCPuGlboAFi7r5gB3YPx9jThJTyqHwD91F4KJGEhhGsrzweg2iecQls1PcL9TQ6oGUOnw8HdkLmMYT1C2V1YzsHyGrOjEkK4EElYOJm/twfenhZ8vVr5VG/6DGorZDpIV+flB6lTjeRVdfOjJ2KC/fjHxYP47Jax+Hl5cMvc37BV13VsnEII58rfYtw3rBBSW29nQ3YJQxrme3c432BqA+NIsewjr1SWNhXCpZUXAJBvDwIgPswFExb9zwPvQPjt7aalmldlSh0LIcTvJGHhZH7eHgT7eaFUK5eRWvueMVc5boRzAxOub/AVUFsOW+Yf97DEiAD+eflQ9h2s4N6P11FVK4U4hXAbR6wQsi23jOo6u3kJC8DSbQD91F52F5abFoMQohVsRsJiX40VgB6umLDwDjCWcd/0KUMjNaH+Xny6RlZBE0L8ThIWThZh9Wl9YbSinbB3mVG/oLUJDuG+epwMoYmw7r0WDx3ZM4z7z+nLwg25TH35Z/aXVDo/PiGE8+VvNqaHBccDNM3vNjNh4dFtAMmWHHblFpkWgxCiFRqmhOyqNBIV8aEuOtV4xDVQX43PxnlcOCyObzblyZQzIUQTSVg42YOT+/HvacNad/D6DwEFgy93akyik1DKGGWx+yco3tfi4TeOT2b2NSeRfbCSC/+9TIZrC+EO8jZDVP+mJPaGrGLCA7yJM/ODR3R/PLBTs3+reTEIIVpmywcPH3aXeeDn5UFYQCvrqXW0bgMh7iRY9V+uOCmOOrvmf7/JKAshhEESFk4WYfVp3ZxBrY2ERdJ4CIp1fmCicxh8OaBh/QetOvy0lCjenzma3NIq3v1VljwVolPT2pgScsgKITvybfSJDmz9NENnaFgpJKh0m0xBE8KVlReANYp8Ww3RQT7m9hstGX4NFO2gV8U6TkoM5YMVe9Famx2VEMIFSMLCVWStgoN7YOClZkciXEloIiSMhXUfGB9eWmFA92DG94lk3qp91NXbnRufEMJ5SrOhusQYYQForcnIs9E72mpuXGHJ1Fu86aP2satA6lgI4bJs+RAQSX5pFVGBrZyebJbUC8A3GFbP5oqRPdhTVMEvu2TamRBCEhauY8M88PSFflPMjkS4mkGXQVEG7F/b6odcflIP8kqr+XxtjvPiEkI4V95m475hRENeaTVl1XX0jjI5YeHhSW1YCv3UXnbkl5kbixDi2MrzwRpFga2ayEAfs6M5Pm9/GHwlbP6CSUmeBPl68sGKlqfDCiHcnyQsXEF9LWz8BPqcA75BZkcjXE3/88DiBRs+bvVDzugXRUp0IH/6aB0jn1jMmc//QLkseSpE55K30biP6gfQlBzoFRVoVkRNvGIH0Neyjx15zS+7LIRwAbYCCIikoLQTJCzAKL5pr8V3/btcOCyORRtzOVBeY3ZUQgiTScLCFez6ASoKYeAlZkciXJFfKPSeABv/B/bWzRf38rDw+W1jufusPpyUGEZGvo3/Lt3t5ECFEA6VvxmC4sAvBKApOWD6lBCMlUKiVDEH8mUUlxAuyW6H8gJq/SIpq67rHAmLyBRISoOVb3L+oEhq6u38slOmhQjR1UnCwhVsmGfM2+t9ltmRCFc18GIo2w+ZP7f6Ib5eHtxxRm9enjaMCf2jefXHXRyUbyqE6DzyNh9VcDPU34twV6j0H50KQFDZdpMDEUI0q/Ig6HrKPEMBiOoMCQuAUTdDWQ6pB9NRCpl2JoSQhIXpasphywLoPxU8O8mLieh4fc4Bbyts+KhND//jmb2xVdexYMN+BwcmhHCKuhoo3NaUGADIyC+jd5TJK4Q0aqirEVmeYXIgQohmlecDUKxCADrHCAswRpSGJeG9+jXiQ/3ZkS/TzoTo6iRhYbZtX0FtOQyS1UHEcXj7Q99zYfPnUFd9wg/vHxNEUmQAX0nCQojOoWgH2Osg6veExa6CcpIiA0wM6hDWSEo9Qomt3ml2JEKI5tiMhEUhwQBEWDtJwsJigZE3QtZKzg7JYkeejLAQoquThIXZ1s8z5ij3GGN2JMLVDbwEqkogY/EJP1QpxeSBMfy6q4hC24knPIQQHaxphRBjSkhJZS1F5TUkRrhIwgLI80smoU5q4wjhksoLACjQRsIi3OoCU8laa8iV4B3I+dUL2F1YTq0s0S5ElyYJCzOVF8HOJTDwIiOjLMTxJI0H/4g2TwuZOCAGu4aXv8/AbtcODk4I4VB5G43VgcJ7A7CnsByAni6UsDgY2IckvQ9dLysQCeFyGkZY7K8zVhUK9e9ECQvfIBg6jf4HFhNaf4DMonKzIxJCmEg+JZtpy+fGkF9ZHUS0hocXpF5gTCOqPvEhkv1iArliZDyzf97DQ59vdEKAQgiHyd8MEX3A0/iQsafI9RIWtuAUfFUtlXk7zA5FCHGk8nyweLG/2pcAbw98vTzMjujEjLoRhZ1rPBfJ8slCdHGSsDDTpk+Nb88aipcJ0aKBl0BdFWz98oQfqpTiyQsGMvPUJN5bvpfP12Y7IUDhjpRSPkqpN5VSmUqpMqXUWqXURLPjcmu5Gw8ruLm7sByloEeYv4lBHa4mvB8AVVnrTY5EiM7H6f2qrQACIjlQUUtYZ5oO0igsifq+5zHNYzGZOblmRyOEMJEkLMxiy4c9S41vzF2h4rvoHOJHQnAP2Pi/Nj1cKcW9Z6cwrEcID3+2EVu1DOUWreIJ7APGA8HAQ8A8pVSimUG5rfJCKMuBmEFNm3YXlhMb7Oda35JGplCnLdhzZcSWEG3g3H61PB+skRSV1xDWmaaDHMLz1LsJUpXE7JhrdihCCBNJwsIsW74AbTcSFkK0llKQej7s/B4qi9t0Ck8PC3+dkkppVR0frNhLnRSzEi3QWpdrrWdprfdore1a6wXAbmC42bG5pdyGEQvdfk9Y7Cksd6npIACBViu7dQyWgi1mhyJEp+P0ftWWDwFRHCyvISygcyYsiBnMBt8RjCv6CGqrzI5GCGEST7MD6LI2fgoRKRDVz+xIRGfT/wJY9i/YttCopN0GQ+JDGNkzjBcW7+DpRdv42/mpXD6yh4MDFe5KKRUN9AE2NbNvJjATIDo6mvT09I4Nrg1sNptLxRm/9zOSgaU7S6nbm47Wmh25FYyO8WxznM5o456Seg7oeMblrnOJ58/V/o7OIG10X8fqVw/tUyMjI1v93Iw+sI+DOoycolKCtIdLPacn8jde438ud1XNYuuHs8iNO8e5gZ0Ad/l3Ku1wHe7QBmeRhIUZynIh82cYf59MBxEnrvswY1rIpk/bnLAA+OMZvbn+rVWE+Hvxz8U7uGBYd3w8XWi4uXBJSikvYC7wltZ665H7tdavAa8BjBgxQqelpXVsgG2Qnp6OS8X58dsQHM8pZ00BoKCsmoqvFzN2cB/STunZplM6o437DlTw/ooeTKn9lbTRw4zK/iZyub+jE0gb3dPx+tVD+9SUlJTW9alaw09ldEseRHm2on9yPGlp/R0feBudyN+40JrMmi/eZUD+Qvpe8Th4uMZHF3f5dyrtcB3u0AZnkSkhZtj8BaBlOohoGwdMCwEY2yuCzX87m2cvGUxuaRUfr85yXIzCLSmlLMA7QA1wm8nhuK/cDYdNB9mUUwJA/xhzEwJHCvLzYpuON37Jl2khQrSFU/rVqmKor6HKJ5yqWjvRQb4OOa0Z+nQL5D91U/AqzTS+qBFCdDmSsDDDpk8hqj9E9TU7EtFZ9b8A7LXGtJB2UEoxrncEg+KCmf3zHrTWDgpQuBullALeBKKBi7TWtSaH5J5qyqFwx2EFNzfllALQP9a1EhaBPp5s0w1TyfKk8KYQJ8pp/aqtAIASSwgAkYE+DjmtGZIjrXxrH05hQC/44Smol2LhQnQ1krDoaKU5sPcXGV0h2ufQaSHtpJTiD6MTyMi38fiXW5jyr6XklkhxK3GU/wD9gCla60qzg3FbeZsADd0GNm3alFNCQrg/wX5e5sXVDItFUeYTQ5UlAPI3mx2OEJ2Rc/rV8nwACnUwQKceYRHg40lsSACfBV8NRRmw4SOzQxJCdDBJWHS0zZ8DGvpPNTsS0Zk5aFpIo3MHxRLk68mbS3ezIbuEV37Y2f4YhdtQSiUANwJDgFyllK3hNs3cyNxQMyuEbMwuZUBssEkBHV+wvzfZPkkNiRYhRGs5tV+1GQmL3HpjVFZnTlgA9Im28r+KIRAzuGGUhQzwE6IrkYRFR9v0KUQPgMg+ZkciOjsHTQsB8PP24Ka0ZAZ0D2LigG68v2IveaUyykIYtNaZWmultfbVWlsPuc01Oza3s389+IVCcBwAJRW17D1QQWp315oO0ijYz4vdHolGwkKmlAnRak7tV8uNKSFZtYEARHXiKSEAvaMD2VlYTv34B+DgHlj7ntkhCSE6kCQsOlJpDuxbLqMrhGM4cFoIwC1pvVhw+zjun9gXDVwzeyVFtmqHnFsI0Uq5643pIA0rSG3abxTcTHXVERZ+XmwnAapLoTjT7HCEEGCMsFAeZFb6YPXxJMDHNVbWaKveUVZq6uzsDT8Fuo+AH5+BOnl/IkRXIQmLjrT1S+O+/3nmxiHcg4OnhTRKCA/g9atGsLPAxl8/l2HeQnSY+jrI23z4CiHZRsHNVBcruNko2M+LdfWJxi85a80MRQjRqDwfAiLIL6slKqhzj64AY4QFwIacUjjtASjZB7+9bXJUQoiOIgmLjrTlC4joA5EpZkci3EXTtJCvHHra8X0iuerkBL7elEu+TA0RomMUbof6amOedoONOSXEBPsSYXXNDx09IwL4sSQKbfGCnN/MDkcIAcYqIQFR5JVWER3YuetXgLGkc3yYH//+PoP6nqdBwimQ/hRUlZodmhCiA0jCoqOUF8Gen6HfFLMjEe6k+zAI6g5bFzj81FeM7EGdXTNv1T6Hn1sI0Yz9a437QxMW2SUuOx0EYHB8CJV2TypC+0HOGrPDEUKAMcLCGkl+WTXRbjDCwtvTwn3n9GVrbhn/+y0bJjwGFYWw9HmzQxNCdABJWHSU7V+BrpeEhXAspaDvZMhYAjXlDj11UqSVsb3Cef2n3azac8Ch5xZCNCN7NXgHQnhvAMqr69hVWM4AFy24CTA43kim7PXtY0wJsdvNDUgIAbYCdEAkeaVVRHXyFUIaTR4Yw9AeITz7zTYqIgfBoMvhl3/DQamdI4S769QJC6VUsFJqRcMyUAMO2f4PpdRPSql3lFKusXD9lvkQHA8xQ8yORLibvudCXaWRtHCwJy8YSKi/Fxe/8gujnlzMhqwSh19DCNEgezXEDgGL8dK8NbcUrXHZJU0BogJ96R7ix291iUbhzQO7zA5JiK5NayjPp8Ynguo6e6dfIaSRUoqHJvcjv6ya13/cDWc8DMoCSx41OzQhhJN16oQFUAFMBj5u3KCUGgx011qPA7YCF5sU2++qy2Dnd8boiobK70I4TMJYYxlEJ0wLSQgP4NNbxvLgpH7U1Nn513c7HH4NIQRGxfvcjdB9eNOmjQ0FNwd0d92EBcCQ+BC+Ke5u/CJ1LIQwV3UZ1FVR6hkKQLSbjLAAGJ4QxqSB3Xj1x53kqwgYczts/B/sW2F2aEIIJ+rUCQutda3WuuCIzWOAbxp+XgSM7diomrHjG6ivkekgwjk8PCFlEmxfBPW1Dj99aIA3N5yaxJWjerB4Sx77DlQ4/BpCdHm5G40CuockLLbsLyXU38vl56APjg9maUkE2tNP6lgIYbZy421xsQoBcNmCvW113zl9qa2388TCLTD2j2CNhq8fkOloQrixDktYKKVuU0qtUkpVK6XmHLEvTCn1qVKqXCmVqZS6sh2XCgUaywaXAGHtOJdjbFkAAZEQP8rsSIS76nsuVJXAnp+cdonpoxNQSvHEl1uoq5c3BkI4VPZq4777sKZNGfk2ekcHolx8ZN5JiWHU48HBoL6QLSMshDCVLR+AooaERbjV28RgHC8hPIBbT+vF52tzWLCtFM54BLJWwpp3zA5NCOEkHTnCIgd4HPhvM/teBmqAaGAa8B+lVCqAUqqbUiq9mVu3Y1ynGGisUBYMmFstsLbKGGHRdzJYPEwNRbix5NPAK8ColeIkMcF+3Ht2Cos25XLJq7/wv9VZTruWEF1Ozm/GN4VBxtQKrTU78m30jrKaHFjLBnYPxurjyVZLMuSuh/o6s0MSousqNxIWhXbjrXCov3slLABuPa0Xg+NDePDTjeT2vNCYGvvtX43lXIUQbqfDEhZa60+01p8BRYduV0oFABcBD2utbVrrpcAXwB8aHpertU5r5pZ7jEstA85s+Pls4GdntKfVdqVDjU2mgwjn8vKDXmfA1oVOHRZ54/hknrxgIAfLa/jTR+vYU+jYlUmE6LKyVxvTQRpGUxTYqimprO0UCQtPDwsnJYaSbouH2goo3GZ2SEJ0XQ0jLPbXGwmLEH/XqD3vSF4eFv7v0sHU1Nn58//WY5/0vLFS2jcPmh2aEMIJPM0OAOgD1Gmttx+ybR0wvjUPVkotBIYAKUqpV7XWc5RSeUqpn4C9wLPHeexMYCZAdHQ06enpbWvBcaRsfY1IjwB+3qvRWe0/v81mc0qcrkTa2DZRJNPf9gW/zX+N0uC+Dj33oWKBmf00DyyFt75axvj45t8Myd9RiFaqKoHC7TDw0qZNGXk2AHpFBZoV1Qk5OTmcD7fH8oAPRh2L6FSzQxKiayovABT7a/0J8i3Fy6NTl6s7pqRIKw9O7sdDn23knZ3RXH3KXfDj0zD4CmPUqRDCbbhCwsLK7zUnGpUArXqXprWe1My2P7fysa8BrwGMGDFCp6WlteZhrWevhxXXQr+JjD/9LIecMj09HYfH6WKkjW1UOQS2/YthftmQdpNjz30ErTXPr11CsXcEaWlDmj1G/o5CtFJjocpD61cUGAmL3tGuP8ICYHRSOH/XMdR6BuCVvRqGTjc7JCG6Jls++IdTWGEnLMD9poMcatqoHizZksffv9rCmX+8he4bP4Yv74abfwEv91kdRYiuzhXSrjZ+rznRKAgoMyEWx8paBRVFkDLR7EhEV+AXAj1PNZY31dqpl1JKMSopjF93FaGdfC0h3F5jocrYoU2bduTZCPT1JCqwc1T479stCC8PT7L8U2HvcrPDEaLrKi+AgEgOVtQQ6uYJC6UUj18wEK3h6cV74Nz/gwO74Id/mB2aEMKBXCFhsR3wVEr1PmTbYGCTSfE4zraFYPGEXme2fKwQjtD3XOPFusD5c8hH9wxjf0kV67NKnH4tIdxa9moISwL/3xe12pZXRq8oq8uvENLI29NC35hAVpMC+ZuhstjskITommz5YI2kyFbjlgU3j9Q9xI8bxiXx+doc1ngONkZ3/fxPSZwK4UY6cllTT6WUL+ABeCilfJVSnlrrcuAT4G9KqQCl1FjgfKDzr0+0fREkjDG++RaiI/Q5x7jfttDpl0pLicLf24ML/v0zf3hzOT9sl+rcQpwwrWHvrxA3smlTeXUda/cWMyIh1MTATlxqbDBfl/YEtDHCUAjR8crzISCK/SWVdAvuGtMibk5LJjLQh8cWbEaf/XcIjodPZ0J15x+sLYTo2BoWDwGPHPL7dOBRYBZwC8Zyp/kYq4jcrLV2iREWdrudrKwsystPcDUEex0M+xv4hcKWLQ6LJzg4mC0OPJ8rkjaeuICAAOLi4rAEd4eYwUaybNzdDjt/c+LD/Em/J423f8nk0zXZ3Pzuar6/J43ooK7xBkkIhyjKgIpCSDi5adMvO4uoqbeTlhJlYmAnbkD3IJ5YkYj280Dt/QV6y+hCITqcrYAavwgOVtQSH+pvdjQdIsDHkz9PSOHe/61n/jYb5134GsyeCIv+Aue/ZHZ4Qoh26rCEhdZ6FkZyorl9B4CpHRXLiSgsLEQpRUpKChbLCQxIseVDaR1E9QdPx81BLisrIzCwc1SNbytp44mx2+1kZ2dTWFhIVFQUpEyC9KeM9citkQ65xrFEBflyz9kpXDIijrOe/5FH52/i7xcMItgNl1ETwikylxn3PcY0bfphewH+3h6MSOxcIywGdg+mAl9Kg/sSvE+GYwvR4WrKobacYksIAPFhfubG04EuGh7HnGV7+MdXW5nwp/H4jr0Tlj5vjDztd67Z4Qkh2sEVali4tOLiYqKjo08sWQHGMnWevg5NVgjRHIvFQnR0NCUlDbUk+pwDaNjxTYfFkBAewM1pySzckMuQx75h7FPfsbGwvsOuL0SntfcX8I+ACKOMk9aa9O35jEkOx8fTw+TgTkyf6EA8LYodvqnGlJD6WrNDEqJrseUDUGAPBiCui4ywAPCwKB4+tz/ZxZW88sNOSPuLMeJ0/h1Qlmd2eEKIdpCERQvq6+vx8jrBb4vtdVBjA99g5wQlxBG8vLyoq6szfokZDIGxsP2rDo3hzjN787+bx3D3mX3QWvPpjpoOvb4QnVLmMmM6SENxzV2F5ew7UMn4TjYdBMDXy4P+sUH8WNUL6iph/3qzQxKiayk3akll1xrLIceHdp0RFgAnJ4dz3uBYXvoug7X7K+DC16GmAj6+VhKoQnRikrBohROu0l5VatxLwkJ0kMP+jSoFKedAxndQW9WhMQxPCOX2M3pzzdie7Cyxk5EvBa+EOKbSHCjOPGw6SPo24wNHWh/nTudyllE9w/ikMM74Zd+v5gYjRFfTkLDIrLLi5+VBmJsva9qcx84fQFSgD3d+sIbyoGQ470XIXArfPtLyg4UQLkkSFs5QVWosZ+rlPkPxrFYru3btOub+m266iccee6xd10hPTycuLq7Vx6elpfHGG28AMHfuXCZMmNCu6x8qNTWV9PR0AGbNmsX06dMddu4nn3yS2267zWHna1afiVBbDnuWOvc6xzB1aHc8FNz23hr+8sl66u3alDiEcGmN9SsOKbiZvi2f5MgA4sM65+vHqJ7hZNWFUGWNN1Y/EUJ0nIYpITsq/IgP8+s0yyI7UrC/F/932RAyD1Tw6PxNMOhSGHUT/PoybPjY7PCEEG0gCQtH03aoLgWfoKYhvs6SmJiIn58fVqu16easD8I2m42kpKRj7n/llVd4+OGHnXLt1pg2bRrffNNyzYYZM2bw0EMPtXjcpk2bSEtLa3dczSVhHnjgAV56yclVq3ueaiTMOmB50+ZEBvpweg9PbNV1vL9iHx+u3GdKHEK4tMxl4B0I0QMBqKypZ/nuA51udZBDnZQYhlKwx3+AkbDQkqwUosPYjFoNm0p8ulT9iiONSgrnlrRk5q3KYuGG/TDhcWMk2xe3Q55LLEIohDgBkrBwtJpy0PUdNh1k/vz52Gy2ppvTPwg3o77efYorNtWB6Oy8fCH5dGN5U5M+MEzr58NP957GqJ5hPP31VpkeIsSR9v4C8SPBw1iw65ddhdTU2UlL6ZzTQcD4drNvtyCW1vSB8nwo3GF2SEJ0HbY88A9nb0ktcV2sfsWR7jyzD4PjgvnLJxvILquDS+YYXyZ+MA0qD5odnhDiBEjCwtGqSgAFPuYuyzlnzhzGjh3LXXfdRUhICElJSSxbtow5c+YQHx9PVFQUb731VtPxM2bM4KabbuKss84iMDCQ8ePHk5mZ2bRfKUVGRkbTsTfffDOTJk0iICCA77///qiRC59//jlDhgwhKCiI5ORkFi1aBMDs2bPp168fgYGBJCUl8eqrr7a6Td9++y19+/YlODiY2267DX3IB/E5c+ZwyimnAEaV/bvuuouoqCiCgoIYOHAgGzdu5LXXXmPu3Lk8/fTTWK1WpkyZAhgjVf7xj38waNAgAgICqKurIzExkcWLFzedv6qqissuu4zAwECGDRvGunXrmn1uGp+fhx56iPLyciZOnEhOTk7TCJicnBxmzZrF9ddf33T8F198QWpqKiEhIaSlpbFly5amfYmJiTz77LMMGjSI4OBgLrvsMqqqWlmXos85UJoNueYVvlNK8fjUAShg0otLWbqj0LRYhHApFQcgf/MR00EK8PPyYGTPMBMDa79RPcP4oKhhRN6u780NRoiuxJaPPSCKsqo6ugX7mh2Nqbw8LLxw+VDq7ZqZb6+iwiccLn0bSrKMpEUH1vgSQrSPJCwcSWsjYeETCBbzl6Nbvnw5gwYNoqioiCuvvJLLL7+clStXkpGRwbvvvsttt92GzWZrOn7u3Lk8/PDDFBYWMmTIEKZNm3bMc7/33ns8+OCDlJWVNSUKGq1YsYKrrrqKZ555huLiYn788UcSExMBiIqKYsGCBZSWljJ79mzuuusufvvttxbbUlhYyIUXXsjjjz9OYWEhycnJ/Pzzz80e+8033/Djjz+yfft2SkpKmDdvHuHh4cycOZNp06Zx7733YrPZmD9/ftNj3n//fb788kuKi4vx9PQ86pyff/45l1xyCQcOHODKK69k6tSp1NYev+J0QEAAX331FbGxsU0jYGJjYw87Zvv27VxxxRX885//pKCggEmTJjFlyhRqan5fYWPevHksWrSI3bt3s379eubMmdPi8wVAn7MBBdsWte54J+kdHcjXd51KRIA3r/107DoownUppW5TSq1SSlUrpeaYHY9byGzovxLGAg3LmW4r6JTLmR5pVM8wMmojqQpMgJ2SsBCiOU7pV215VPkaI7S6BXXthAVAYkQAL14xhM37S7nno3XY40bCBa8Y/e8nN4DdfUYIC+HOjv5kJo7p0fmb2JxTeuwDtB1qK8DTByz5bbpG/9ggHpmS2urjp06detgH7GeeeYYbbrgBgJ49e3LNNdcAcNlll/HEE0/w17/+FR8fHyZMmIC3tzcZGRkMGTIEgMmTJ3PqqacC8MQTTxAcHMy+ffuIj48/6rrnn38+Y8cab7R9fQ9/UXzzzTe59tprOeusswDo3r17077Jkyc3/Tx+/HgmTJjATz/9xLBhw47bzoULF5KamsrFF18MwJ133slzzz3X7LFeXl6UlZWxdetWRo4cSb9+/Y57boA77rij2XY2Gj58eNO17777bp577jl+/fVXxo0b1+K5j+fDDz9k8uTJTc/VPffcwwsvvMCyZcuaamjccccdTYmOKVOmsHbt2tad3BoFcSOM5U3T7mtXnO0VFejLRcPjePn7DPJKq/D19ODTNVn84eREftt7kFB/L3pFmTsqSRxXDvA4cDbQtccZO0rGYmN4ctxJAOwuLGfvgQpuGNfT5MDar3GESEbgCAbs+dpYTtDjBJcHF8L9Ob5fteVhCzPeT0nCwnB632gemNiPJxZu4YWoHdx11sXG1JmvH4BF98PEp51ec04I0T4ywsKR7A31D1TH5YE+++wziouLm26NyQqA6Ojopp/9/Pya3XboCItDP7BbrVbCwsLIyclp9rrH+3C/b98+kpOTm9331VdfMXr0aMLCwggJCWHhwoUUFrY8TSAnJ+ewayqljhnD6aefzm233catt95KVFQUM2fOpLT0OImmFtpz5H6LxUJcXNwxn5sTkZOTQ0JCwmHnjo+PJzs7u2lbt27dmn729/c/7G/Wot4TIGcN2AraHWt7XTgsDruGj1dn8ej8Tcyav5mfMwq58Z3VPPXVNrPDE8ehtf5Ea/0ZUGR2LG5Ba8hYYhTHbfgg/8P2huVMO3HBzUbhVh96R1n5riYVamyQtdLskIRwOQ7vV7WGsjyKLaEAREnCosn143pyyfA4XliygwXrc+DkW+Hk22DFa/DzP80OTwjRAhlhcQJaHPlQuB3sdojq2zEBOdi+fb+v5GCz2Thw4MBRUxgaHW+prPj4eHbu3HnU9urqai666CLefvttzj//fLy8vJg6dephtSiOJSYm5rD4tNaH/X6kO+64gzvuuIP8/HwuvfRSnnnmGR577LFjxt3S0l+HXstut5OVldX03Pj7+1NRUdG0Pzc3t2llkJbOGxsby4YNG45q16GjUtql91nw/ROwcwkMvtwx52yjnhEBjOsdwbPfbGuqAzp3eSYHymvILCo3NTbhOEqpmcBMMBKkjcsDuzKbzdahcfqX72NkyT62RU9hf8N15/9WRaSfYuf6FRzde7ZfR7cxzqead3Liud3LQuZ3s9nTs6blB7VTR7fRDNLGrufQPjUyMvKYz41nrY1T6qvZVmhMc9ixfiVZnq45csCMv/GEcM3aEAt3f7CGwt1bSAw6g35R64lePIttewvYHzvhhM7nLv9OpR2uwx3a4CySsHAUe52xQog1uuVjXdTChQtZunQpI0eO5OGHH2b06NEtjjxoznXXXceECRM499xzOe2009i/fz9lZWV0796d6upqIiMj8fT05KuvvuKbb75hwIABLZ5z8uTJ3HbbbXzyySecd955vPzyy+Tm5jZ77MqVK7Hb7QwbNoyAgAB8fX2xWIzBRNHR0ezadeJ1FFavXt107RdffBEfHx9Gjx4NwJAhQ3jvvfdITU3l22+/5YcffmDEiBFN1ysqKqKkpITg4KNXjrn00kt56qmnWLJkCaeeeiovvPACPj4+jBkz5oRjbFa3wRAQCTu+MT1hAfDK9OE8On8TuwrKsVXX8fUmYwm2vQcqsNs1FotrvrkSrae1fg14DWDEiBHaEcsDO1t6erpDljFutV9eBiBl0s2khPTAbtfc+eO3nNU/hrS0wU65ZEe3sSw0h+/fX0NF7GAS63eT2AHX7vC/owmkjV3PoX1qSkrKsfvUgu3wM9SEJGE96MnEM0/ruCBPkFl/46Ejqzn/pZ/5zyY7X9w2luhxY+GDK0nZ/jIpfXrBiGtbfS53+Xcq7XAd7tAGZ5EpIY5S3bBko09Qh152ypQpTStQWK1WLrjggjaf68orr+TRRx8lLCyM1atX8+6777bpPCNHjmwqqBkcHNy04khgYCAvvvgil156KaGhobz33nucd955rTpnREQEH330Effffz/h4eHs2LGjqYbGkUpLS7nhhhsIDQ0lISGB8PBw/vznPwNGMmXz5s2EhIQwderUVrfp/PPP58MPPyQ0NJR33nmHTz75BC8vYyj3Cy+8wPz58wkJCWHu3LmHnbdv375cccUVJCUlERISctQ0kpSUFN59911uv/12IiIimD9/PvPnz8fb27vVsR2XxQK9zjKGn7tAcakAH0+evngwH988hnG9I5q2V9fZyS+rNjEyITpQxhKISIGQHgBszy+juKKWUUnhJgfmOKOSjDoWW/1HQPZqWUZQCGezGV8A7KuxEh3kY3IwrinC6sMbV4+grKqOmW+vokp7wmVzjVXVFtwFy1u/cp0QouPICAtHqSoF5QHeAR12yT179hxz34wZM5gxY0bT77169Tpq6kVWVtZhv0dERPDKK680e74jlxA90pHbLrjggmaTJ7feeiu33nprs9dIS0sjKyuLsrKyZvefc845bN++vdl9h7b3jDPOYP365pfy7N2791FFK5t7Hg/dNmvWrGbP1WjEiBFs2rTpmPv/+9//Hvb7rFmzDmvjsZ6r5mJrKZZm9T4T1r0HWaugx6gTf7yTjOwZzus/7SYy0IeCsmoyi8q7/DJsoguorTQq1B/yTd7yXQcAY3UNdxEV6EtyZABfV/VjuLbD7p+gf+sS1EKINmhIWOyssspr6XH0iwnin5cN4cZ3V3Pvx+t54fIhqEvfgY+vga/uhfoaGHO72WEKIQ4hIywcQWtjhIVPoFQaFq4n+XRQFsj41uxIDjMyMYwQfy+uO8VYFWFDdgnv/JqJ3d5yTRPRsZRSnkopX8AD8FBK+SrVgdWF3cmen6GuCnqd0bRp2c5CYoN9iQt1rwVYRieFMy+3G9o70OX6HyHM5vB+1WasTrfd5k+0FNw8rgmp3bhnQgpfrMvh5e8zwNMbLpkD/afCNw/Bj89AK+qrCSE6hiQsHKG2Euy14Nux00GEaBW/UIgbadSxcCHB/l6s/esErjulJx4WxXPfbOfhzzayfPcB/rFoK5/8ltXySURHeQioBO4Hpjf8/JCpEXVWGYvB0xcSjCltuSVVLNmSz9kDurVYpLezGZ0UTnE1FHcfD9sWGUWphRCNHNuv2nLRHt5sL/MgLtTfMRG6sVvSkpk6JJZnv9nOZ2uyjRWbLnoTBl0G3z0OX9wOdc4vFiyEaJl8Q+YI1Q1LZnZw/QpHam6ah3Ajvc+C7x6DsjwIdK3CsF4eFmJDfNl3oBIwVg5ZsH4/3p4WUmODSekWaHKEQms9C5hlchidn9aw7UtjOVMvYzTFW7/swa4114zpaXJwjjcmORyl4FevUUwsXwDZqyB+pNlhCeESHN6v2vKp84tElyuSIztuenJnpZTiqYsGsb+kirvmrcVWXcf00Qkw9RWjvtCPz8DBPXDZO8YXP0II08gIC0eoLgVPPyM7K4Qr6n2Wcb9ziblxHENCmPHmKtjPiwXr9wPg5+XBze+uliVPhfvIXQ/Fe6HfFACqaut5b/leJvTvRo9w9/tGNNzqw/Aeocwp6A0WT9j6pdkhCeG+bHmUexmFe3tGSMKiNXy9PHjr2pGcnhLFQ59t5OXvM9BKwekPwQWvwr7l8MaZUOSMxaaFEK0lCYv2alzOVKaDCFfWbRBYu7nctJBGg+ODSYoM4Oa0ZACG9gjh9atGcKCihgv+vYySilqTIxTCATZ/YRRnTpkMwNebcimprOWqkxNMDsx5zuwfzfL9dqrjxsCWL2ReuBDOYsvnoMUYCSAJi9bz9fLglT8MZ+qQWJ75ehtPLtxiFJoffDlc9QVUHIA3zoCd35kdqhBdliQs2suk5UyFOCFKQa8zjRfc+jqzoznKPRNSWPTHUzmrfzQWBRcOi2NkzzDevPokDpTXsGjTfrNDFKJ9tDY+sCeOhQDjW9APV+4jPsyP0W60nOmRzuxnTEH7LfA0OLAL9q8zOSIh3JQtj1x7MJGBPgT6yojfE+HlYeH5S4dw9ckJvP7Tbu6et47y6jpIOBluWALWaHjnQlg8C+rlCxQhOpokLNqruuOXMxWiTXqfCVUlkLXS7EiOopTC29NCcqSVb+8ez7SRPQAY1iOExHB/5q/7PWFRb9fkl1aZFaoQbZO7AQq3G1XogfVZxSzbWcQlw+OxWNyr2OahkiMD6B1l5T95/Y1pIRv/Z3ZIQrif+looLySz2kqSjK5oE4tFMeu8VO4+qw+frc1m0os/sTrzIIQlwQ3fw7CrYOn/wX/PMWpbCCE6jCQs2kNrqJLlTEUnkXSakVxz8eUFkyOtTR/glFJMGRzLsp2FFJRVA/De8kzGP5NOSaV8yyE6kfUfgsULUi8gs6icGbNX0j3EjytH9TA7MqdSSnHpiHh+zKqnPO5U2PSprBYihKOV7Qc060oDGJYgBSLbSinFHWf05oMbRlNXr7nklWU89802aj184bwXjaVPC3fAK+OIzP/J7HCF6DIkYdEedQ3Lmcp0EFPMmTOHU0455Zj7P/30U+Lj47FaraxZs4bU1FTS09M7LkBX4xcC8aNcto7FsZw/JBa7hheWbAdg+e4DVNbWsymnxOTIhGglez1s+Ngofusfxnsr9lJWVcu7148iwupjdnROd8Gw7nhaFN96ngol+yDzZ7NDEsK9lOYAkGMPY2xyhMnBdH6jksL56s5xXDA0jn99l8GF/17GxuwSSL0AbvoJIvuSuvlZmHc1lMqUVSGcTRIW7VHVsJyprznLLiYmJrJ48WJTrt0Z3HPPPbz00kvYbDaGDh3Kpk2bSEtLA2DWrFlMnz7d3ADN0PssY2h6Wa7ZkbRar6hArj+lJ+/+upcfthewIdtIVGzOKTU5MiFaadf3YMuFQZcCsDG7hJRugV2mMF6E1YezU7vxxK5eaJ8gWPOu2SEJ4V5KsgDI0eGEBXibHIx7CPL14rlLB/OfacPIOljBuf9ayu3vr2F3fQRcs5BdPafBtq/g5ZGw4nUjMS2EcApJWLRHdVnDcqby4tAadXUdW+wxMzOT1NTUDr2my2tc3jSjcyW67jk7hfgwP575eiuZRRUAbJKEhegsVs8B/3BImYTWmo3ZpQzsHmx2VB3qpvHJFFRZ2Bx2Fmz+3KinI4RwjNJsAPbrMIL8PE0Oxr1MHBhD+p9P49bTklm8OY8zn/+Bv3y+lXXRF8Mtv0D34bDwHnjzLNi/3uxwhXBLkrBoK3sd1NhccjnT6upq7rzzTmJjY4mNjeXOO++kurq6af/TTz9NTEwMsbGxvPHGGyilyMjIaPZc6enpxMXF8dxzzxEVFUVMTAyzZ89u2l9SUsJVV11FZGQkCQkJPP7449gb5ifPmTOHsWPHctdddxEeHs6sWbOYMWMGt9xyCxMnTsRqtTJ27Fhyc3O58847CQ0NpW/fvqxb93sV+aeeeork5GQCAwPp378/n376aavab7Vaqa+vZ/DgwSQnG0tlNo5IWbRoEU8++SQffvghVquVwYMHt+l57pSiB0BgTKebFuLr5cFFw+LYmG0kKYJ8PY3hmUK4urJc41u4IVeCpw/7DlRSUlnLgC6WsBgYF0xaSiR/zzvJmE65fp7ZIQnhPkqyqfGwYsOfYD9ZIcTRgv28+PPZffnh3jSmj+rBx6v3ce+PlTy/uo6Kyz6CC9+A4r3wWhosuEumiQjhYJKwaKtqm3HvgvUrnnjiCX799VfWrl3LunXrWLFiBY8//jgAixYt4vnnn2fx4sVkZGS0qqZDbm4uJSUlZGdn8+abb3Lrrbdy8OBBAG6//XZKSkrYtWsXP/zwA2+//fZhCY3ly5eTlJREXl4eDz74IADz5s3j8ccfp7CwEB8fH04++WSGDRtGYWEhF198MQ888EDT45OTk/npp58oKSnhkUceYfr06ezff/wXAh8fH2w24++zbt06du7cedj+c845hwceeIDLLrsMm812WILE7SkFyWfArh863fDFqUO6N/18wdDu7CywUVnTudoguqDf3jYS3MOuBmBjQ+2VrjbCAmDmqUksrehBUfAAWPGaFN8UwlFKsyn1jsSiIMBbRlg4S1SgL4+eP4Dv/pTGsCgPXvwugzOe/5HP7WPQt66Ek66D396BF4fCt49A5UGzQxbCLUivdiK+ut+Y/w9QV2W8CfUOABy4Qki3gTDxqXadYu7cufzrX/8iKioKgEceeYQbb7yRxx57jHnz5nHNNdc0TZWYNWsWc+fOPe75vLy8+Otf/4qnpyeTJk3CarWybds2TjrpJD744APWrl1LYGAggYGB/OlPf+Kdd97huuuuAyA2Npbbb7/9/9u78/CoyrPx4997lsyELAQSSEiAsAnIJggqiCKgqOgPUamAUBWtdrFWbatWXtEi0hd3rYrWFQu4tNal7q9WQQWtKFYERERIAgkQSCD7MpPk+f1xTtIhBhKyzZL7c11zZeZscz9zztyTuec5zwHA5bIOtwsuuIDRo0fX3X/00Ue59NJLAZg1axaPPPJI3XNfdNFFdfdnzZrFkiVLWLduHdOnT2/Ra9Sh9Z8EX6+E3V9Dz9HBjqbJ+iTFMKp3AoVlfsb1T+Kvn2Ux4Z5VXHv6MVwyNh1fVQ3LP8vkknHpeFzOYIerFPgrrC/mA86ApGMA2JhTiNspDEoJzthHwTSuXyKDkuNY5j+TGwrvt8b2GHB6sMNSKvwV5XDQ1Z04rzuiL5McKnp17cSvRnq5oc8Ibn9jM9e9+DUr0rvwx2kLGD72ali9BNb+GdYvg/HXwUm/tL8vKKWaQ3tYNJczClxeWrVY0Up2795Nenp63eP09HR2795dN69Xr1518wLv79y5k9jY2LpbrcTExLpiA0CnTp0oKSkhLy8Pv9//o+fKyclpcPu1kpOT6+5HR0f/6HFt7wiA5cuXM3LkSBISEkhISGDTpk3k5eU1/cVQP9ZvovV3x4dBDaM5Hpo9iscvGc3JAxI5Z3gKCdFu7nx7C7lFFXyybT+L39rCqu/2BztMpSwbX4LS/TDuGgAeeP97nvpkB8f1TOiQRTUR4dKT03niwHH4orvBpw8FOySlIkNhDpn+BLrFRf5Vh0LJCX268vqvT+HuGSPIzC/lvKVr+N37hWRPehB+tRZ6nwwfLIIHhsGHi6E4N9ghKxWWtIfF0Whhz4f2kpqaesiAkzt37iQ1NRWAHj16kJ2dXbfsrl276u737t37kGJBY5KSknC73WRlZTFkyJC650pL+2/XfZHmF3SysrK46qqr+OCDDxg3bhxOp5ORI0dijGn2NlsjrrAXkwQpI2D7KphwY7CjOSq9unaqu//o3NFk5Zcy5f6Pue+9rQxMtn6x3pZbzNnDUoIVolKW6ipY8wAkD4d+E9mwq4A/f7CNc4ansHBaxx0M+IJRadz1zne8FXMhF+x4HLK/hJ5jgh2WUuGrqhJK97HRH8cFZ6Q1vrxqVQ6HMPOEXkwdnsLSVdtZtjaDN7/Zw6Xj0vn19OV0OfC11dvi43utv8NnwrhfQ/KQYIeuVNjQHhZhzu/3U1FRUXerqqri4osvZvHixezfv5+8vDwWLVpUdwnPmTNnsmzZMrZs2UJZWRl33HFHs5/b6XQyc+ZMbrnlFoqLi8nKyuL+++9vtcuFlpaWIiJ069YNgGXLlrFp06ZW2XZycjKZmZl1A4R2OP0nw651/x2LJUylJ8Zw9rAU1mzLq7t6yPf7wrtNKkJsfAkObIeJfwARlq76gXivi7tmjKB7vDfY0QVNpygXM8f04o+7T6LGkwAf3RXskJQKbwezANhputcV7lX7i/O6uXnqYFbdMJHzR6XyzNoMJty9iqU/dKX8wuXwm/XWWEabX4HHxsGKC+Db16HKF+zQlQp5WrAIc+eccw7R0dF1t4ULF7JgwQLGjBnDiBEjGD58OMcffzwLFiwAYOrUqVx77bVMmjSJAQMGMHbsWMAaqLI5Hn74YWJiYujXrx+nnHIKc+bM4YorrmiVtg0ZMoTf//73jBs3juTkZDZu3Mj48eNbZdu1Y2MkJiZy/PHHt8o2w0r/SVDjh6y1wY6kxY7tEc/uwgq+yS4ArB4WSgVVVSV8dCekjKB64Lk8/tF23vs2l3nj+xLn1RH8Lx3Xh2Lj5aPuc6wrFmWuCXZISoWvAzsAyDLJpCd2amRh1dZSE6K5+yfH8e71EzipXyL3/N9Wxt35AUs+97Fr3CL47WY4/TbYtwX+fgncNwjevskaV6wVehArFYn0lJAwlpmZedh5Dz30EA891PD5wfPnz2f+/PkAbNmyBYfDQY8ePRpcduLEiYecQlL/ebt06cLKlSsbXHfevHnMmzfvkGnPPvvsIY+vvPJKrrzyyrrHAwYMqLsCCVhXPPnTn/7U5O0Hqn/qSGDciYmJrFnTgf9J7jXWGoNl+4cw8KxgR9Mig+3BCzdkW1df2L6/BH91DW6n1mNVkPz7MTiYSf75L3D1U5/zecYBzh6awi8m9At2ZCGhd2Inzjsules3j+Wrzq/hfG8BXPkhOPQ9q9RRswsWlfF9GNAttpGFVXsZmBzHU5eNYX3WAZ5ek8FTazJ44pMdTB7UnUvGXcKE667FkfERfP0crH8W1j0O3YfCyIthyHRI6B3sJigVMvS/gw7o1VdfpbKykoMHD/KHP/yBadOmHTKopuoA3F5IH2+NYxHmAq+2kJYQjb/akJVfSnWN4ZNt+1nxWSbVNfqrhWonRbvh43upOeZs5qyKZVNOIff8ZASP/fR4YjyaZ2v99oyBlFa7ebnL5bD7P/DVs8EOSamw5Nv/A0UmmsmjBusVQkLQ6PSuPDp3NGv+MInfTBrAhuwC5i37gskPrGHJD2msHXUPFddtgXPvB3c0vLcAHhwOfzkFVt9pXZ1Qe16oDk4LFh3Q448/Tvfu3enfvz9Op5PHHnss2CGpYOg/CfK2QmFO48uGsB6dvcR5rS+CZw61rjizdW8J177wHy55eh23/nMz67P0WuiqHRgDb/4Waqp4NvYqtuYW8/CcUVw0plfHHui3AX2SYrh8fB9u2jaEvG4nwfsLoWhPsMNSKuxU7vuBLJPMkLTOwQ5FHUGPztH87sxBrL15Mn+ePZLUhGieWZPB3Kc+Z+Q967h043CeGvwkO2Z/jJlyB7hjrILFX06BB0fAOzfDtn+F/dhjSjWH/tzTAb377rvBDkGFgv6Trb87VsGo1hkoNRhEhMEpcXyReZAzjk3mxXW7uOkfGyj1VXPR6J68tD6bPYXlwQ5TdQRfPwffv0vxabdz57/8XDAqjcmDkxtfr4O68azBfJl1kMtyL+ZN93zktV/CT1/VU0OUOgqOgxlkmVT6dNXxK8KBx+Vk+sg0po9Mo7Syis8z8vn4+7y6S7MvBvokDmfq8DM571QXg4vXIlvfhi+fgc8fA4cL0sZA3wnWrecJVq9ZpSKYFiyU6qi6D4HYZOu0kDAuWAAMTonni8yDDEqJY+WVJ/Hoqh/ondiJ300ZyEvrs9lbWBHsEFWky90Mb90AfU7loZLTqTa7+O0ZA4MdVUiLcjl4cNZIpjxQxIspV3PxjvtgzX1hd7llpYLGV0Z0aTbbzRgm6ICbYSfG42Ly4OS6wnZOQTmrt+7j3U17eeLjHTy22tCzSy/OGb6QqXMfYHj1FlxZn0DGx/DJvfDx3dZ4ZD1PsC4PnTYG0kZDfMPj0ikVrrRg0QTGGO3Oq0Ja/QFGm0QE+k2CH96Hmpqw/lVz1gm9iI92kRgTRVKsh6fnnVA3L9bjYm9RBTvzy/C6HR36kpKqjRTvpWTZT/DVRHN71TW8+3k25x2XSm/9AtGo9MQYrp7Yn/n/qmZE2lkM/XAxJA2CIecFOzSlQl7Jro3EUkOGsy/xegWisJeWEM3ck9KZe1I6B0t9vL8ll3c27mHZ2gye+NjgdAipCRPokziVgcNqOEm+Y1D51yQfXI/n04eRmiprQ3Gp0HO0VbzoMRKSh0JMN+v/PqXCkBYsGuF0OvH7/URFRQU7FKUOy+/3N2/g1P6T4JsXIXcj9Diu9QNrJ8PSOjPsMOfvJsd7yC2q4Bcr1+N1O3j16ta5NK7qWEoqq4htaNDM0jxKnj4fKc/nltglZJXGMGVIDDedPaj9gwxTV08cQG5RBReuu5jX47IY+PLPkKgXYcDpwQ5NqZBV4a8mY/O/GQ507dcBL88e4brERDFzTC9mjulFYbmfj77fz7bcYrLyy8jKL+WlXaU8XdEdOBM4Ew8+JsTt4ZToLEbIdvpmfk3Cljfqtmc6JSLdh1i9a7sfaxUxuunnlAoPWrBoREJCArm5uaSlpeEI41+gVeSqqakhNzeXzp2bMeBWv4nW3+0fhnXB4kh6dI5m54EytuUWU1Vj+G5vEYNT4oMdlgojH2zJ5Zcr1/Po3NFMGZLM5t2FfLWzgJkDDDw/E1fBDm6PvZX7r5tHdJQz2OGGnSiXgyUXjmBsv0Quf+VGljkWMeD52TgufAIZdkGww1MqJE2+dzW/KP2Evk4v1/1kSrDDUW2oc7Sb845L/dH0gjIfmXYBIyu/jMz8vryeP5KH80vJK/GRQDHHOnYyWHYytDSHobty6Jv1JV7z33G9TnR1puK7ATiT+uNOGgBd+9m3vtCpa3s2U6nDCuuChYh0Bt4HhgBjjTGbRORE4M+AH8gBLjXG+Jv7HElJSWRnZ7N169ZWibmlKioq8Hoju0u7tvHoxcTEkJSUdPQrxqVY1/3evgpO+W2rxRNKkuO9fLo9j9orm764bhcLzxsa3KBUWHlmbQb+asP8V74hKXYMP1+xnkGlX3Ju1FJcVHGjYwELfnaVFitaaPrINEb0nMr8ldHccGAhY/4xj33bvqD7ebeDU7u7KxVod2EFo6J+YJPpy9gYT7DDUUGQ0CmKkZ2iGNkr4UfzSiqryMovZWd+GTkF5Ww6WM67B8vJOVBCTcEuevozOEZy6F2VS5+cXNL3fECa/P2QbVQ4OlHqTcEX0wMTl4ozoRfepN7EdEvH1aWX9T9kVKyeaqLaXFgXLIAy4FzgnoBpu4DJxphyEVkCTAf+0dwncDgc9O7du2VRtqLVq1czatSoYIfRprSN7az/JFj3JPjKICryzrnv0dlbV6w4pnss//w6hz9OG6Lj0qgjKvNV4XU5ycwvZe0P+Uwfmcq7m/Zy9aNvcHPUy8yIWs0edzpLe97BdVMm0UtH6G8VfZNiWH7NVP72aT92rrqFCzcsZc+WN6masoReJ5wb7PCUChnxlDBMMvm/bpcFOxQVgmI9LoamdmZoasO9bwvL/ewuKOeDtevw9B3MNyWVHCwsxBzIwl2URWxpFvGVe0gszqNHyR5S922imxT+aDuVeChyJlDi7kqZO5FKTyL+6G7UxCRBTHdcsV1xxyYRHZ9EdOckYmPjiYt243Zqr3XVdGFdsLB7TuwP/OJhjAm8kLsPqGnvuJQKK/0nwWePwM5PYcAZwY6m1SV3tnqyOAQuPrE3i978ll0Hyg8ZEHHXgTJW/juLm84ejNOhhYyObn3WQS57Zh1dY6LwVdXgcgi3jXWz2Psh0RuexSEGxl5Lj0n/w63u6GCHG3GiXA4umXAsxSf+jX++9iyjttxD77fm8NV7oykYPo9hp82ge+eYYIepVIuJSFfgaayBCPKA+caY5xtbb8ueIk5xbMIhhinnXtTWYaoI1DnaTedoN7lJLiaO7hkw59DxUCr81eSX+sguquDrgmLK8nfhO7ATU5iDu2wfnsp8OvnyifMfIL5iJ8lFG+lKkfU52YBK4yafWIqIpdgRT7kzFp8zFp8rlipXLNVRcRhPHHjiEW88juh4XN54nN5YXN5OuDwxuLwxeD0evC4nHrcDr9tJkc9QUlmF1+XApcWQiNNuBQsRuQaYBwwHXjDGzAuY16yE3cjzpdvbW9yS7SgV8XqfDM4o67SQCCxYpNhXBUlPjGFMny4AbMwpPKRgcerdqwD46dj0ul/K9xVX8M7GvVw6Ll17Y4SA97/N5YH3v+e2aUMY2y/xiMtWVdfw2Y58BqfE0y3O6iptjOGz7fls21fC5xn5xHnc/OzUvmTmlbJ0fQU50VnMPSmd9zbv5YaXNpAcI5wWm8Gxvo2cHbeeuGc3gDhhxCyYNB8SQqfnXaSK87qZPvsq8gtmse6NexiwYzldv/oNu9f/kXc6jUf6T2LgiWfSp2dPHFpoVOFpKdaPa8nASOAtEdlgjNnc0MK+auhz81sAPOX+hDJPdzr1HttesaoOyOt2kpYQTVpCNPTuAvQGjjx4ud/vo+hALmUFe6ksyqeyJJ/qknxqyg5A2QEcFQU4Kw/SxVdIStV+oqoyifaVEm3KcDbxd+ZK46KCKMrxUG6iGICXrWuiKDdRVIqHSvHgc0RT5fDid3qpcnipdkVT44ymxuVBnF4cLjcOtweHKwqn24PD7cHp8uCKsm9uD+4oD1FRXtyeaKI81n1vlBuP24XX7cDjduJ1O4hyOvR/xTbUnj0sdmMVD84C6v8kddiELSIpwIsNbG+2MWZvQ08kIvHACmBeS8avUKpDiOoEvU6CjI+CHUmb6GH3sDimeyyDUuJwO4WNOYWcO6IHN7/8DTvyShtc7653tvLyV9kckxzLyf2bMT6IOmpZ+aV8k13IppxCMvNL6d8tluN6JbApp5BHV28H4LJn1nHbtCGM75/E5zvy2JRTwKCUOCp9VeQUlFNc4WPrnmK27i0kyulgSI9YBnaLobKqmnc27saLj55xhp3lpdy0fhUxUk5/RxE7Xn+dle+V0tm3l3+699KncjdS5rMCSxkOZy6G4TMhLjmIr1DHlJgQT+Ild2CqbiX781fwr1/JpAPv4d30BmyC3SaRDEc6B+jK81/8g3Jvd3zebji9sbij43BHx+GNiaNTp1hioz3Eej24XC7E6UIcTpxOJw4RHAJi/3WI4HQIYt+vne9wyI+WddrT6i+r/7yqIxGRGGAGMMwYUwKsEZHXgUuAmxtaJ7eshu4YJjv+wxnO/1B14o3gDOvO0ioCud1RJCT3IiG519GtaAz4SqGyCCqKqCovpLKkgMrSAqory6iqLKWmsowafxmmshTjLwd/GQ5/ORTsJ9njwFFdjrOqBFd1Hs6aCqJqynH7K3C34tfBGiP4ceHDhR8nB3Hjx0UVTqrtWw0OqsVJNQ7rPk5qxHHI/Bo59L6/2vDWmofrlq0RF6ZuXSdGrG0hDhDB4MCIAA6MWNMN2PN/PA/kv+uKAwK2Vbd87bJ18+WQZWpvYj+/OGqfy4FI2xVt2i3LGWNeARCRMUBd36PGErZdlJjY1OcRERdWgeN2Y0xojJSpVKjrexqsWgxlByJuVOgUu2AxMDkOj8vJoJQ4NuUUYozhjQ27KfVVN7henNdKj19kHNSCRTvZ/+SFTC7/D6cDTjHwg4HPDJMwXB9lndZjjIG3DQ4x9AFmAWxoYGO1Y9rm2Tfg7tppfqxPv3qfgP4aNxWxycT0HIJ0Ow96ngDp4yHmyD06VPsQVxQ9x8+G8bOhqpJ9337C3m/X4Ni3mb5lOxheuY340mJouAZ5RFXGQTUODE37Z6v+csa+1f9tsOnba5oxCKWrmrhwmDoBmtVG1/wMPN6wGktmIFBljPk+YNoG4LTDrZBi9rPVcyluqWavpw8pp1zf1jEq1X5EwBNr3eJT6z6mm3IC4OrVqxkyceLhF6ipBn+5fSuDaj9U++ybH6orMVU+/L5K6+avoMpXSZWvgmp/JdV+n/W3ykeNv5Kaah/Gb61j7O1ItQ8x1YipwWGqcZlqxFTjsKeJqcKBH4eprJvuoHaZGozx4QZreo1d6jCBf61PqVB3RRtsU4xp34aLyGKgZ+0pISIyClhrjOkUsMwNwGnGmGlN2N7bWL0ysoDHgWrgQWCjvchjxpi/HWbdnwM/tx8OAsKhwJFE3b/fEUvbGBm0ja0r3RjTrZ2eq1WIyH6s3Bzq9FiNDNrGyNAh8qqInAq8ZIxJCZh2FTDXGDMxYFrg/6rDgE3tGWcbiYTjOBLaANqOUBIJbQAYZIyJa80NhkI/sligqN60QqBJDTXGnNPA5BVNXPcJ4ImmLBsqRORLY8yYYMfRlrSNkUHbqMKlwNIR9qO2MTJoGyNKCRBfb1o8UBw4IfB/1Uh5bSKhHZHQBtB2hJJIaANY7WjtbYbCMKpNSthKKaWUUkpFiO8Bl4gcEzDtOKDBATeVUqqjCoWChSZspZRSSinVYRhjSoFXgEUiEiMi44HpNLGXsFJKdRTtVrAQEZeIeAEn4BQRr4i4NGEftbA6haWZtI2RQduowkVH2I/axsigbYwsV2NdOW8f8ALwq8Nd0tQWKa9NJLQjEtoA2o5QEgltgDZoR7sNuikiC4E/1pt8uzFmoYh0BZ4BpgD5WFcHeb5dAlNKKaWUUkoppVTIaferhCillFJKKaWUUko1JhTGsFAtICKrReTKYMfRnkRkoohkBzsOpVRk6mh5VXOqUiqUREJOEpF5IrIm2HG0VCTsCwAR6SMiRkRC4QqZ6ihpwSJEiMg1IvKliFSKyLP15p0uIt+JSJmIrBKR9CCFedSa2y4R8YjIMyJSJCJ7ReR37R58Cxyu3XbiNyLyar3lj7Onr27vWI9GC/bnart9x9Vb51V7+sR2aUAD2uIYDff9HCkiMa921JwKmlc1r4b3fm4JEelq79dSEckSkTnBjgnq9tfTdkzFIvK1iEwNmB9WOUlEjhGRChFZGTBtjt2+UhF5TazT2Gvnhdx+EZHZIrLFjmm7iJxqTw+bfSFWceFtETlox/OI2IUGERkpIuvtdqwXkZEB64mI3CUi+fbtLhGRoDWknnD7DDtSHq+33GX2vigSkWwRuVsCCkMikikiPhFJqrfef+z29WksFi1YhI7dwGKssTzq2Dv3FeBWoCvwJfC3do+u+ZrbroXAMUA6MAm4SUTObod4W0uD7bbtB8aJSGLAtMuwrpgT6lpynH4PXBqwTiIwDuv1CKa2OkbDeT9HikjMqx01p4LmVc2r4b2fW2Ip4AOSgbnAYyIyNLghAeACdgGnAZ2BBcDf7S+c4ZiTlgJf1D6wX+PHgUuwXvsy4NF6y4fMfhGRKcBdwOVAHDAB2BGG++JRrIFoewAjsY6vq0UkCvgnsBLoAvwV+Kc9HeDnwPlYV5scAUwDftGegTci3D7DjhRvoE7A9UAScBJwOnBDvWUygItrH4jIcHu9JtGCRYgwxrxijHkNa9DRQBcCm40xLxljKrCSynEiMrj+NkSkh4h8IyI3tnnATdSCdl0G3GGMOWiM2QI8Ccxr6DlE5FoR+VZEerZFG5rjCO0G68PtNWA2gIg4gVnAc+0VX3O18Dh9DphltxesxPUq1usRNG14jIbtfo4UkZhXO2pOBc2raF6FMN7PzSUiMcAM4FZjTIkxZg3wOtaX6KAyxpQaYxYaYzKNMTXGmDexvpiMJsxykojMBgqADwImzwXeMMZ8bIwpwfrCf6GIxB3tfhGRe0RkjYh0bqs2ALcDi4wx/7b3R44xJocw2xdAX+DvxpgKY8xe4F1gKDARq0j2oDGm0hjzECDA5IB23GeMybbbfd8R2jHD/uV/WBu24xDh9hnWSLyByz1mjPnEGOOzX/fngPH1FltBQHEda18tb2osWrAIfUOBDbUP7MvAbren1xGRvsBHwCPGmHvaNcLmOWy7RKQLVlV1Q8DyG6jXZgARuQ0rGZ1mjAmnc+yW89837lnAJqxKZrhqynG6G/gWONN+fClHkayCoDWO0Ujbz5EiEvNqR8+pEHnvN82rmlcBBgJVxpjAX1obfP8Gm4gkY8W7mTDKSSISDywC6p8KUb8N27G+WA6kiftFRBwi8iTWL/5nGmMKW78FdV9wxwDdROQHu2v+IyIS3UA7QnZf2B4EZotIJxFJA6by36LFN+bQK0Z8ExDrIe3k8O24HKsnyhnGmE2tH36zRUpum4CVAwL9G4gXkWPtY3U2Vk+ZJtGCReiLBeont0Ksrl61hgCrgD8aY8LlGr5HaldswOP682qJiNyP9U/aJGNMsLu/HhVjzKdAVxEZROj/g9kUTTlOwU7GdlU/wRjzWXsE10wtPUYjcT9HikjMqx06p0JEvt80r2peBet1Kao3raHjIKhExI31y+pfjTHfEV456Q7g6Qa+hDfWhsb2ixt4Aes0jGnGmLJWi/jHku3n+wlwKtapFKOwTtMJp30B8DFWoaEIyMY6heU1Gs+J9ecXArEih4xjcT1wIzDRGPNDawfeEpGQ20TkCqzC2b0NzK7tZTEF2ALkNHW7WrAIfSVAfL1p8UBxwOO5WDv9H+0VVCs4UrtKAh7Xn1crAetctSVtVa1uByuAa7DOF3y1kWVDXVOOU7DOoZyM1e4V7RBXS7T0GK0VSfs5UkRiXtWcaomk95vmVc2r0PTjIGhExIG1T3xY+wXCJCeJNWjjGcADDcxurA2N7ZcBwHTgdmNMW5+mVW7/fdgYs8cYkwfcD5zTSKwhsy+g7lh6FyuvxWCNi9AFq0dEY695/fnxQEm9Hhk3AktDuAdh2OY2ETkfWAJMtY+/+lYAc7B66RxVMUYLFqFvM9bgMUDduYz9ObSrzUIgD3g+4DzWUHfYdhljDgJ7Aufb9wPbfBD4f8AyEal/nlS4WAFcDbzdxlX39tCU4xS7ne8AvyL0/7Fu6TFaK5L2c6SIxLyqOdUSSe83zauaV8EadM8lIscETDvc69Lu7F+vn8b6hX+GMcZvzwqXnDQR6APsFJG9WIMFzhCRrxpoQz/Ag7VPmrJftmANgPmO/at5m7Ff02wg8Mt57f1w2Rdg9UbpjXUqZqUxJh9YhlV42QyMqNdjYkRArIe0k4bfJ2cCC0RkRlsE3wrCMrfZg7Q+idWTaGNDyxhjsrDGuDkHqyDVZFqwCBEi4hIRL+AEnCLiFeuSMK8Cw+zBYbzAbVjnb30XsLofuAirErncrk6GhBa0azlWQulid3O9Cng2cNvGmNVYv4K+IiIntk+LmuYI7a5jjMnAGvn4lmDE2BwtPE5r/Q/W+Y+Z7Rb4EbTlMQrhuZ8jRSTm1Y6aU0HzKppX64Tjfm4ue7yBV4BFIhJjf2GcTugUpx4DjsX6olIeMD1cctITWF/eR9q3vwBvYY0h8BwwTUROtb/kLwJeMcYUN3W/GGNewHp//ktE+rdRG2otA34jIt3FGpvit8CbhM++wP5lPgP4lZ1HErAGaPwGWA1UA9eKdSnW2t48Hwa043cikiYiqcDv67cDq4BxNrBURM5rq3Y0JNw+w5oSr73cZKz3ygxjzLpGNvszYLL9/mk6Y4zeQuCG9WueqXdbaM87A/gOq7vXaqBPwHqrgSvt+17gX1hvTkew29TCdnmwLqNTBOQCvwuYNxHIDnh8rr3M8cFub2Ptrh97vXWuBFYHO/a2Pk4b2GY21rmE4damJh+j4bafI+UWiXm1o+bUI7U93N9vmlc1rzbh9eyKdQ5/KbATmBPsmOy40u19W4HVHb/2Nrc193V75iT72F0Z8HiO/ZqXYl1Ss2tT9gtWt/c1AY+vArICX4M2iN2NdUnQAmAv8BDgDbd9gVU4Wo3VsyMP+DuQbM8bBay32/EVMCpgPQHuBg7Yt7sBsef1sY9Vl/14jN2Oqe34fllIGH2GHSHe3ljv8972cquAKg7NAe8EbCcTa4DT+tt32dts9D1RuxOVUkoppZRSSimlQkZIdHFVSimllFJKKaWUCqQFC6WUUkoppZRSSoUcLVgopZRSSimllFIq5GjBQimllFJKKaWUUiFHCxZKKaWUUkoppZQKOVqwUEoppZRSSimlVMjRgoVSTSAiJSLSrw22u0RErm/isveJyK9aOwallAoGzatKKdW6NK+qSCTGmGDHoFS7E5FMIBmoBkqBd4BrjDEl7RhDN+BrYIAxpjxg+hxgkR3fDuAMY8x+EekBrAP6G2N87RWnUko1heZVpZRqXZpXldIeFqpjm2aMiQWOB8YAC+ovICKuNnz+ecDb9ZJ/F+Ap4OfGmDhgNlAGYIzZA3wHnNeGMSmlVEtoXlVKqdaleVV1aFqwUB2eMSYHq2I9DEBEjIj8WkS2AdsCpg2w70fb3d2yRKRQRNaISLQ9b6yIfCoiBSKyQUQmHuGppwIf1Zs2ECgzxnxox7bFGFMaMH81cG5L26yUUm1J86pSSrUuzauqo2rLapxSYUFEegHnAK8ETD4fOAkob2CVe4GhwMnAXnu5GhFJA94CLgHeBU4HXhaRwcaY/Q1sZziwtd60aKDiCOFuAWY00iSllAoqzatKKdW6NK+qjkp7WKiO7DURKQDWYFWO/zdg3hJjzIHA7m8AIuIArgCuM8bkGGOqjTGfGmMqgZ9idZl72xhTY4x5H/gS68OlIQlAccC2LwfeBFLtineBiJxSb51iez2llApFmleVUqp1aV5VHZr2sFAd2fnGmH8dZt6uw0xPArzA9gbmpQMXici0gGluYNVhtnUQiKt9YIxZJiIZwEpjTM/DrBMHFBxmnlJKBZvmVaWUal2aV1WHpj0slGrY4S6fk4fVBa5/A/N2ASuMMQkBtxhjzJ2H2dY3WOcAHo1jgQ1HuY5SSoUCzatKKdW6NK+qiKcFC6WOgjGmBngGuF9EUkXEKSLjRMQDrASmichZ9nSviEwUkcNVn98GTjvKEE7DGnBJKaUiguZVpZRqXZpXVSTRgoVSR+8GYCPwBXAAuAtwGGN2AdOB/wH2Y1Wwb+Tw77PlwDkBIzb/BSu5p4pIScDtVHt+D2AI8FpbNUwppYJE86pSSrUuzasqIogxh+tJpJRqayLyv8A+Y8yDTVj2PmC7MebRNg9MKaXClOZVpZRqXZpXVTBpwUIppZRSSimllFIhR08JUUoppZRSSimlVMjRgoVSSimllFJKKaVCjhYslFJKKaWUUkopFXK0YKGUUkoppZRSSqmQowULpZRSSimllFJKhRwtWCillFJKKaWUUirkaMFCKaWUUkoppZRSIUcLFkoppZRSSimllAo5/x/7CFNpxmeTlQAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 1080x288 with 3 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.rcParams['font.size'] = 12\n",
    "\n",
    "df = df.dropna()\n",
    "x = df['Price_adj']\n",
    "log_x = np.log(x)\n",
    "log_bins = np.arange(8, 19, 0.05)\n",
    "bins = np.exp(log_bins) #np.arange(1e3, 1e8, 1e5)\n",
    "y, _ = np.histogram(x, bins=bins, density=True)\n",
    "log_y, _ = np.histogram(log_x, bins=log_bins, density=True)\n",
    "\n",
    "# Normal fit - Baseline\n",
    "mu_singleFit, std_singleFit = norm.fit(log_x)\n",
    "pdf_single = norm.pdf(log_bins, loc=mu_singleFit, scale=std_singleFit)\n",
    "\n",
    "# Plot\n",
    "norm_y = sum(np.diff(bins)*y)\n",
    "norm_log_y = sum(np.diff(log_bins)*log_y)\n",
    "norm_single = sum(np.diff(log_bins)*pdf_single[1:])\n",
    "\n",
    "# Dif bins\n",
    "dif_bins = np.diff(bins)\n",
    "dif_log_bins = np.diff(log_bins)\n",
    "#=====================================================================\n",
    "# PLOTS\n",
    "#=====================================================================\n",
    "fig, (ax1, ax2, ax3) = plt.subplots(1, 3)\n",
    "fig.set_figwidth(15)\n",
    "\n",
    "# Plot 1\n",
    "ax1.semilogy(log_bins[1:], log_y/norm_log_y/dif_bins*dif_log_bins)\n",
    "ax1.semilogy(log_bins[1:], pdf_single[1:]/norm_single/dif_bins*dif_log_bins)\n",
    "plt.setp(ax1, ylim=[1e-12, 1e-4])\n",
    "plt.setp(ax1, xticks=np.log(10**np.array([4, 5, 6, 7, 8])), xticklabels=['10k', '100k', '1M', '10M', '100M'])\n",
    "ax1.grid()\n",
    "ax1.legend(['Empirical distribution', 'Log-normal fit'])\n",
    "plt.setp(ax1, xlabel='Price (£)')\n",
    "plt.setp(ax1, ylabel='pdf')\n",
    "\n",
    "# Plot 2\n",
    "ax2.plot(log_bins[1:], log_y/norm_log_y/dif_bins*dif_log_bins)\n",
    "ax2.plot(log_bins[1:], pdf_single[1:]/norm_single/dif_bins*dif_log_bins)\n",
    "plt.setp(ax2, ylim=[1e-11, 6e-6])\n",
    "plt.setp(ax2, xticks=np.log(10**np.array([4, 5, 6, 7, 8])), xticklabels=['10k', '100k', '1M', '10M', '100M'])\n",
    "ax2.grid()\n",
    "ax2.legend(['Empirical distribution', 'Log-normal fit'])\n",
    "plt.setp(ax2, xlabel='Price (£)')\n",
    "plt.setp(ax2, ylabel='pdf')\n",
    "\n",
    "# Plot 3\n",
    "ax3.plot(bins[1:], y/norm_y)\n",
    "ax3.plot(bins, pdf_single/norm_single/np.exp(log_bins))\n",
    "plt.setp(ax3, xticks=[0, 2e5, 4e5, 6e5, 8e5, 1e6, 1.2e6], \n",
    "         xticklabels=['0', '200k', '400k', '600k', '800k', '1M', '1.2M'])\n",
    "ax3.axis([0, 1.2e6, 0, 6e-6])\n",
    "ax3.grid()\n",
    "ax3.legend(['Empirical distribution', 'Log-normal fit'])\n",
    "plt.setp(ax3, xlabel='Price (£)')\n",
    "plt.setp(ax3, ylabel='pdf')\n",
    "plt.tight_layout()\n",
    "plt.savefig('figures_mixture/lognormal_fit.png', dpi=600)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## EM Algorithm"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Log Transformation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "x = df['Price_adj']\n",
    "### Run the EM algorithm\n",
    "## Initialize the parameters\n",
    "KK_vec = np.arange(6, 7)\n",
    "\n",
    "# Parameters\n",
    "w_vec = []\n",
    "mu_vec = []\n",
    "std_vec = []\n",
    "w_ini_vec = []\n",
    "mu_ini_vec = []\n",
    "std_ini_vec = []"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "===========================================\n",
      "Mixture with 6 components\n",
      "[1, -4345968.634045105, inf]\n",
      "[2, -4561142.752665209, 0.04717548436614283]\n",
      "[3, -4677573.910375801, 0.02489135606223656]\n",
      "[4, -4761019.682703069, 0.01752686984900095]\n",
      "[5, -4828835.309894625, 0.014043889020732737]\n",
      "[6, -4887251.656693353, 0.011952801063297795]\n",
      "[7, -4938957.65514636, 0.01046901027773133]\n",
      "[8, -4985323.820098668, 0.009300532247349636]\n",
      "[9, -5027157.514330835, 0.008321540376030017]\n",
      "[10, -5064990.490991956, 0.007469505960259191]\n",
      "[11, -5099205.211596792, 0.006709814409316905]\n",
      "[12, -5130101.277128316, 0.006022505962849053]\n",
      "[13, -5157936.177906341, 0.005396519037450155]\n",
      "[14, -5182951.3166292375, 0.004826427491733671]\n",
      "[15, -5205386.634308244, 0.004310019457755051]\n",
      "[16, -5225485.672313755, 0.0038463483139954276]\n",
      "[17, -5243493.514417743, 0.003434321422248882]\n",
      "[18, -5259650.712175641, 0.0030719145894033168]\n",
      "[19, -5274186.172111279, 0.0027559626189341577]\n",
      "[20, -5287311.137114098, 0.0024823515511861456]\n",
      "[21, -5299215.302163264, 0.0022464014708566244]\n",
      "[22, -5310065.197849173, 0.0020432697681949433]\n",
      "[23, -5320004.439879939, 0.0018682770180150768]\n",
      "[24, -5329155.250590265, 0.0017171221854180886]\n",
      "[25, -5337620.687107154, 0.001585994399590318]\n",
      "[26, -5345487.138687189, 0.001471606118570984]\n",
      "[27, -5352826.801465474, 0.0013711750913136048]\n",
      "[28, -5359699.962458478, 0.0012823779392776903]\n",
      "[29, -5366157.014123997, 0.0012032916011445708]\n",
      "[30, -5372240.178371306, 0.0011323328900669297]\n",
      "[31, -5377984.952031568, 0.0010682018844422373]\n",
      "[32, -5383421.302217368, 0.001009831830096815]\n",
      "[33, -5388574.646027783, 0.0009563463715240004]\n",
      "[34, -5393466.649170534, 0.00090702389779367]\n",
      "[35, -5398115.875193564, 0.0008612682888849171]\n",
      "[36, -5402538.312908189, 0.0008185851646916964]\n",
      "[37, -5406747.805238593, 0.0007785627297662053]\n",
      "[38, -5410756.398655732, 0.0007408563834318437]\n",
      "[39, -5414574.628767919, 0.0007051763756104083]\n",
      "[40, -5418211.7546072975, 0.0006712779057197069]\n",
      "[41, -5421675.951648219, 0.0006389531709042741]\n",
      "[42, -5424974.4715629835, 0.0006080249652888372]\n",
      "[43, -5428113.775090402, 0.0005783415118940171]\n",
      "[44, -5431099.643094906, 0.0005497722746258005]\n",
      "[45, -5433937.26986467, 0.0005222045505569517]\n",
      "[46, -5436631.341883773, 0.0004955406849730401]\n",
      "[47, -5439186.104670285, 0.0004696957848746249]\n",
      "[48, -5441605.419764624, 0.00044459583297816127]\n",
      "[49, -5443892.813550451, 0.0004201761247270126]\n",
      "[50, -5446051.519271882, 0.0003963799669893727]\n",
      "[51, -5448084.513357526, 0.0003731575897288264]\n",
      "[52, -5449994.54695982, 0.0003504652318156886]\n",
      "[53, -5451784.1734563215, 0.00032826436989465043]\n",
      "[54, -5453455.772529614, 0.00030652106536059934]\n",
      "[55, -5455011.571337275, 0.000285205409248958]\n",
      "[56, -5456453.663198323, 0.00026429104873997846]\n",
      "[57, -5457784.024152769, 0.00024375478189654152]\n",
      "[58, -5459004.527694096, 0.000223576209753141]\n",
      "[59, -5460116.957927387, 0.00020373743673681656]\n",
      "[60, -5461123.021367006, 0.0001842228119898611]\n",
      "[61, -5462024.357555112, 0.00016501870535585062]\n",
      "[62, -5462822.548655455, 0.00014611331289528825]\n",
      "[63, -5463519.128153566, 0.0001274964874785172]\n",
      "[64, -5464115.588775476, 0.0001091595908284653]\n",
      "[65, -5464613.389720304, 9.109536381181689e-05]\n"
     ]
    }
   ],
   "source": [
    "np.random.seed(1)\n",
    "log_x = np.log(x)\n",
    "for KK in KK_vec:\n",
    "    # INITIALISATION\n",
    "    w = 1/KK*np.ones((KK, 1))  #Assign equal weight to each component to start with\n",
    "    mu = np.random.normal(loc=log_x.mean(), scale=log_x.std()/KK, size=KK)#\n",
    "    std = log_x.std()*np.ones(KK)/KK\n",
    "\n",
    "    # Initial parameters\n",
    "    w_ini = w.copy()\n",
    "    mu_ini = mu.copy()\n",
    "    std_ini = std.copy()\n",
    "    # Parameters\n",
    "    sw = False\n",
    "    QQ = -np.inf\n",
    "    epsilon = 1e-4\n",
    "    max_iter = 100\n",
    "    i = 0\n",
    "    # x = df_noout['Price_adj']\n",
    "    print(\"===========================================\")\n",
    "    print(\"Mixture with {} components\".format(KK))\n",
    "    while((~sw) & (i < max_iter)):\n",
    "        i+=1\n",
    "        ## E step\n",
    "        L = np.zeros([KK, len(x)])\n",
    "        v = np.zeros([KK, len(x)])\n",
    "        for k in range(KK):\n",
    "            L[k, :] = norm.logpdf(log_x, loc=mu[k], scale=std[k])\n",
    "        Lmax = np.amax(L, axis=0)\n",
    "        for k in range(KK):\n",
    "            L[k, :] -= Lmax\n",
    "        denom = (w*np.exp(L)).sum(axis=0)\n",
    "        for k in range(KK):\n",
    "            v[k, :] = w[k]*np.exp(L[k, :])/denom\n",
    "\n",
    "        ## M step\n",
    "        for k in range(KK):\n",
    "            w[k] = v[k,:].mean()\n",
    "            mu[k] = (v[k,:]*log_x).sum()/v[k, :].sum()\n",
    "            std[k] = np.sqrt((v[k,:]*(log_x-mu[k])**2).sum()/v[k,:].sum())\n",
    "\n",
    "        ##Check convergence\n",
    "        QQn = 0\n",
    "        for k in range(KK):\n",
    "            QQn += (v[k, :]*(np.log(w[k]) + norm.logpdf(log_x, loc=mu[k], scale=std[k]))).sum()\n",
    "        rel_error = abs(QQn-QQ)/abs(QQn)\n",
    "        if(rel_error < epsilon):\n",
    "            sw=True\n",
    "\n",
    "        QQ = QQn\n",
    "        print([i, QQ, rel_error])\n",
    "\n",
    "    ## ASSIGN Results\n",
    "    w_vec.append(w)\n",
    "    mu_vec.append(mu)\n",
    "    std_vec.append(std)\n",
    "    w_ini_vec.append(w_ini)\n",
    "    mu_ini_vec.append(mu_ini)\n",
    "    std_ini_vec.append(std_ini)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### BIC for mixture"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "# n = len(x)\n",
    "# BIC_vec = []\n",
    "# for index, KK in enumerate(KK_vec):\n",
    "#     LL = np.zeros(n)\n",
    "#     for k in range(KK):\n",
    "#         LL += w_vec[index][k]*lognorm.pdf(x, loc=mu_vec[index][k], scale=std_vec[index][k])\n",
    "#     LL = np.log(LL).sum()\n",
    "#     BIC_vec.append(-2*LL + (3*KK-1)*np.log(n))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Store and plot results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "with open('EMfit_LogScale.pickle', 'wb') as f:\n",
    "    pickle.dump([KK_vec, w_vec, mu_vec, std_vec], f)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### No transformation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "x = df['Price_adj']\n",
    "### Run the EM algorithm\n",
    "## Initialize the parameters\n",
    "KK_vec = np.arange(6, 7)\n",
    "\n",
    "# Parameters\n",
    "w_vec = []\n",
    "mu_vec = []\n",
    "std_vec = []\n",
    "w_ini_vec = []\n",
    "mu_ini_vec = []\n",
    "std_ini_vec = []"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "===========================================\n",
      "Mixture with 6 components\n",
      "[1, -36051070.66555333, inf]\n",
      "[2, -35575849.8257936, 0.013357961709608407]\n",
      "[3, -35306239.79036588, 0.007636328224941378]\n",
      "[4, -35168907.97662805, 0.0039049211829124074]\n",
      "[5, -35100674.77568613, 0.0019439284679844588]\n",
      "[6, -35062128.22731208, 0.0010993784554134125]\n",
      "[7, -35035059.4184294, 0.0007726206072436282]\n",
      "[8, -35011893.844969004, 0.0006616486832437202]\n",
      "[9, -34989671.84047764, 0.0006351018264097872]\n",
      "[10, -34967525.91719405, 0.0006333282868232589]\n",
      "[11, -34945450.67152252, 0.0006317058514721757]\n",
      "[12, -34923684.17535287, 0.0006232588767083917]\n",
      "[13, -34902456.603622735, 0.0006081970667913826]\n",
      "[14, -34881917.07921315, 0.0005888301483815127]\n",
      "[15, -34862128.14693682, 0.0005676340868499061]\n",
      "[16, -34843079.18813082, 0.0005467071008032383]\n",
      "[17, -34824701.54171546, 0.0005277187054523998]\n",
      "[18, -34806880.9080733, 0.0005119859400566394]\n",
      "[19, -34789466.99835528, 0.0005005512076067454]\n",
      "[20, -34772281.60942785, 0.0004942266694047025]\n",
      "[21, -34755126.53253923, 0.0004935984587065763]\n",
      "[22, -34737792.65489288, 0.000498991914039055]\n",
      "[23, -34720071.383203976, 0.0005104042412042289]\n",
      "[24, -34701768.984793566, 0.0005274197525327848]\n",
      "[25, -34682723.4932734, 0.0005491348314632848]\n",
      "[26, -34662822.54579042, 0.0005741294569041849]\n",
      "[27, -34642019.2960363, 0.0006005207022240826]\n",
      "[28, -34620342.98380056, 0.000626114889904052]\n",
      "[29, -34597901.308159865, 0.0006486426861794059]\n",
      "[30, -34574873.40194717, 0.0006660300948896047]\n",
      "[31, -34551494.31964694, 0.0006766446071461768]\n",
      "[32, -34528033.708807595, 0.0006794655912699745]\n",
      "[33, -34504772.19738355, 0.0006741534559618444]\n",
      "[34, -34481978.9369646, 0.0006610194983477931]\n",
      "[35, -34459892.979228675, 0.0006409177692234454]\n",
      "[36, -34438710.1014483, 0.0006150891748842542]\n",
      "[37, -34418575.61802432, 0.0005849888632065897]\n",
      "[38, -34399582.78650502, 0.0005521238916522312]\n",
      "[39, -34381775.75027318, 0.0005179207834166821]\n",
      "[40, -34365155.627515234, 0.000483632983889113]\n",
      "[41, -34349688.34797278, 0.0004502887882348552]\n",
      "[42, -34335313.057657816, 0.00041867363465785627]\n",
      "[43, -34321950.226527154, 0.0003893377573962288]\n",
      "[44, -34309508.90458345, 0.00036262022806294713]\n",
      "[45, -34297892.82993694, 0.00033868187483433634]\n",
      "[46, -34287005.28767691, 0.0003175413591440578]\n",
      "[47, -34276752.75592885, 0.0002991103568374933]\n",
      "[48, -34267047.4632041, 0.00028322523950079047]\n",
      "[49, -34257809.02745552, 0.00026967386446614494]\n",
      "[50, -34248965.361238606, 0.0002582170329421451]\n",
      "[51, -34240453.018675745, 0.0002486048463850306]\n",
      "[52, -34232217.137733735, 0.00024058859257851235]\n",
      "[53, -34224211.10338515, 0.0002339289669643496]\n",
      "[54, -34216396.0291489, 0.0002284014432611127]\n",
      "[55, -34208740.129656255, 0.00022379951625306184]\n",
      "[56, -34201218.03668112, 0.00021993640598018038]\n",
      "[57, -34193810.09558194, 0.00021664567588327704]\n",
      "[58, -34186501.667691156, 0.0002137810988040364]\n",
      "[59, -34179282.45598682, 0.0002112160111503919]\n",
      "[60, -34172145.86556546, 0.00020884232583576502]\n",
      "[61, -34165088.40633261, 0.00020656932447850153]\n",
      "[62, -34158109.14242763, 0.0002043223140918068]\n",
      "[63, -34151209.1908302, 0.00020204120910841686]\n",
      "[64, -34144391.27009482, 0.0001996790829114292]\n",
      "[65, -34137659.29903924, 0.00019720072183659144]\n",
      "[66, -34131018.044351526, 0.0001945812070147343]\n",
      "[67, -34124472.81539794, 0.0001918045441754527]\n",
      "[68, -34118029.20396184, 0.00018886235771644661]\n",
      "[69, -34111692.86619716, 0.00018575266227726132]\n",
      "[70, -34105469.34372704, 0.0001824787223245983]\n",
      "[71, -34099363.92055333, 0.00017904800769701957]\n",
      "[72, -34093381.51226489, 0.00017547125052063293]\n",
      "[73, -34087526.58394172, 0.00017176160636806745]\n",
      "[74, -34081803.09314253, 0.0001679339201493191]\n",
      "[75, -34076214.4544325, 0.00016400409492382668]\n",
      "[76, -34070763.52204552, 0.00015998855979415985]\n",
      "[77, -34065452.58747229, 0.00015590383129640984]\n",
      "[78, -34060283.3890098, 0.0001517661612925454]\n",
      "[79, -34055257.13058643, 0.00014759126334297482]\n",
      "[80, -34050374.507477485, 0.0001433941088628426]\n",
      "[81, -34045635.73683703, 0.0001391887840510461]\n",
      "[82, -34041040.59128145, 0.0001349883985848269]\n",
      "[83, -34036588.43405976, 0.00013080503735899194]\n",
      "[84, -34032278.25462815, 0.0001266497470243167]\n",
      "[85, -34028108.70370457, 0.00012253254977772274]\n",
      "[86, -34024078.127110854, 0.00011846247762133498]\n",
      "[87, -34020184.59791211, 0.00011444762116259525]\n",
      "[88, -34016425.94653518, 0.00011049518790834949]\n",
      "[89, -34012799.78869053, 0.00010661156585684798]\n",
      "[90, -34009303.551036365, 0.00010280238902621964]\n",
      "[91, -34005934.49461243, 9.90726022972865e-05]\n"
     ]
    }
   ],
   "source": [
    "np.random.seed(1)\n",
    "for KK in KK_vec:\n",
    "    # INITIALISATION\n",
    "    w = 1/KK*np.ones((KK, 1))  #Assign equal weight to each component to start with\n",
    "    mu = np.random.normal(loc=x.mean(), scale=x.std()/KK, size=KK)#\n",
    "    std = x.std()*np.ones(KK)/KK\n",
    "\n",
    "    # Initial parameters\n",
    "    w_ini = w.copy()\n",
    "    mu_ini = mu.copy()\n",
    "    std_ini = std.copy()\n",
    "    # Parameters\n",
    "    sw = False\n",
    "    QQ = -np.inf\n",
    "    epsilon = 1e-4\n",
    "    max_iter = 100\n",
    "    i = 0\n",
    "    # x = df_noout['Price_adj']\n",
    "    print(\"===========================================\")\n",
    "    print(\"Mixture with {} components\".format(KK))\n",
    "    while((~sw) & (i < max_iter)):\n",
    "        i+=1\n",
    "        ## E step\n",
    "        L = np.zeros([KK, len(x)])\n",
    "        v = np.zeros([KK, len(x)])\n",
    "        for k in range(KK):\n",
    "            L[k, :] = norm.logpdf(x, loc=mu[k], scale=std[k])\n",
    "        Lmax = np.amax(L, axis=0)\n",
    "        for k in range(KK):\n",
    "            L[k, :] -= Lmax\n",
    "        denom = (w*np.exp(L)).sum(axis=0)\n",
    "        for k in range(KK):\n",
    "            v[k, :] = w[k]*np.exp(L[k, :])/denom\n",
    "\n",
    "        ## M step\n",
    "        for k in range(KK):\n",
    "            w[k] = v[k,:].mean()\n",
    "            mu[k] = (v[k,:]*x).sum()/v[k, :].sum()\n",
    "            std[k] = np.sqrt((v[k,:]*(x-mu[k])**2).sum()/v[k,:].sum())\n",
    "\n",
    "        ##Check convergence\n",
    "        QQn = 0\n",
    "        for k in range(KK):\n",
    "            QQn += (v[k, :]*(np.log(w[k]) + norm.logpdf(x, loc=mu[k], scale=std[k]))).sum()\n",
    "        rel_error = abs(QQn-QQ)/abs(QQn)\n",
    "        if(rel_error < epsilon):\n",
    "            sw=True\n",
    "\n",
    "        QQ = QQn\n",
    "        print([i, QQ, rel_error])\n",
    "\n",
    "    ## ASSIGN Results\n",
    "    w_vec.append(w)\n",
    "    mu_vec.append(mu)\n",
    "    std_vec.append(std)\n",
    "    w_ini_vec.append(w_ini)\n",
    "    mu_ini_vec.append(mu_ini)\n",
    "    std_ini_vec.append(std_ini)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Store results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [],
   "source": [
    "with open('EMfit.pickle', 'wb') as f:\n",
    "    pickle.dump([KK_vec, w_vec, mu_vec, std_vec], f)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## PLOT"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [],
   "source": [
    "with open('EMfit_LogScale.pickle', 'rb') as f:\n",
    "    [KK_vec, w_vec, mu_vec, std_vec] = pickle.load(f)\n",
    "\n",
    "KK_log = KK_vec[0]\n",
    "w_log = w_vec[0]\n",
    "mu_log = mu_vec[0]\n",
    "std_log = std_vec[0]\n",
    "\n",
    "with open('EMfit.pickle', 'rb') as f:\n",
    "    [KK_vec, w_vec, mu_vec, std_vec] = pickle.load(f)\n",
    "\n",
    "KK_lin = KK_vec[0]\n",
    "w_lin = w_vec[0]\n",
    "mu_lin = mu_vec[0]\n",
    "std_lin = std_vec[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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