VAR.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "309b0df9-b634-4710-9759-5566b1e891e8",
   "metadata": {
    "tags": []
   },
   "source": [
    "# Modelo VAR\n",
    "\n",
    "Os modelos de vetores autorregressivos se diferem dos modelos AR, ARMA e ARIMA por não serem unidirecionais, isto é, as variáveis preditivas e _target_ se influenciam em ambos os sentidos.\n",
    "\n",
    "Os modelos VAR costumam ser escritos matematicamente em forma de matriz. Uma equação VAR de grau $p$ com $k$ séries corresponde a:\n",
    "\n",
    "$$\n",
    "\\begin{bmatrix}\n",
    "y_1\\\\\n",
    "y_2\\\\\n",
    ".\\\\\n",
    ".\\\\\n",
    "y_k\n",
    "\\end{bmatrix}\n",
    "=\n",
    "\\begin{bmatrix}\n",
    "c_1\\\\\n",
    "c_2\\\\\n",
    ".\\\\\n",
    ".\\\\\n",
    "c_k\n",
    "\\end{bmatrix}\n",
    "+\n",
    "\\begin{bmatrix}\n",
    "\\phi_{11} & . & . & \\phi_{1p}\\\\\n",
    "\\phi_{21} & . & . & \\phi_{2p}\\\\\n",
    ". & . & . & .\\\\\n",
    ". & . & . & .\\\\\n",
    "\\phi_{k1} & . & . & \\phi_{kp}\n",
    "\\end{bmatrix}\n",
    "\\begin{bmatrix}\n",
    "y_{1, t-1}\\\\\n",
    "y_{2, t-1}\\\\\n",
    ".\\\\\n",
    ".\\\\\n",
    "y_{k, t-1}\n",
    "\\end{bmatrix} + \\cdots + \n",
    "\\begin{bmatrix}\n",
    "\\phi_{11} & . & . & \\phi_{1p}\\\\\n",
    "\\phi_{21} & . & . & \\phi_{2p}\\\\\n",
    ". & . & . & .\\\\\n",
    ". & . & . & .\\\\\n",
    "\\phi_{k1} & . & . & \\phi_{kp}\n",
    "\\end{bmatrix}\n",
    "\\begin{bmatrix}\n",
    "y_{1, t-p}\\\\\n",
    "y_{2, t-p}\\\\\n",
    ".\\\\\n",
    ".\\\\\n",
    "y_{k, t-p}\n",
    "\\end{bmatrix} +\n",
    "\\begin{bmatrix}\n",
    "\\epsilon_1\\\\\n",
    "\\epsilon_2\\\\\n",
    ".\\\\\n",
    ".\\\\\n",
    "\\epsilon_k\n",
    "\\end{bmatrix}\n",
    "$$\n",
    "\n",
    "## Modelo VARMA\n",
    "\n",
    "Seria similar ao modelo ARMA, mas com implementação em vetores para múltiplas séries.\n",
    "\n",
    "## Modelos VAR/VARMA em Python\n",
    "\n",
    "Serão utilizados dois *datasets* do [Kaggle](https://www.kaggle.com/datasets/prakharprasad/time-series-data-1?select=M2SLMoneyStock.csv), que serão combinados, resultando em um único *dataset* com dados mensais de Janeiro/1995 a Dezembro/2015.\n",
    "\n",
    "Lista de passos\n",
    "\n",
    "- Análise Exploratória dos Dados\n",
    "- Testar para estacionariedade\n",
    "    - Se não for estacionária, tirar a diferença em grau 1\n",
    "    - Repetir até ter dados estacionários\n",
    "- Dividir em dados de treino e teste\n",
    "- Testar a ordem P\n",
    "- Aplicar modelo com a ordem P\n",
    "- Realizar previsões nos dados de teste\n",
    "- Se necessário, desfazer a transformação do começo\n",
    "\n",
    "## Análise Exploratória dos Dados\n",
    "\n",
    "### Importação das bibliotecas"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "34d749e3-4acd-4fba-88a1-033c8f6802bc",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "%matplotlib inline\n",
    "\n",
    "from pmdarima import auto_arima\n",
    "from statsmodels.tsa.statespace.varmax import VARMAX,VARMAXResults\n",
    "from statsmodels.tsa.api import VAR\n",
    "from statsmodels.tsa.stattools import adfuller\n",
    "from statsmodels.tsa.seasonal import seasonal_decompose\n",
    "from statsmodels.tools.eval_measures import mse,rmse\n",
    "\n",
    "import warnings\n",
    "warnings.filterwarnings(\"ignore\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "1265dc90-81f9-422e-aa5d-0577b96658b9",
   "metadata": {},
   "outputs": [],
   "source": [
    "money_df = pd.read_csv('dados/M2SLMoneyStock.csv', \n",
    "                       dtype={'Money':'float'}, \n",
    "                       index_col=0, \n",
    "                       parse_dates=True)\n",
    "spending_df = pd.read_csv('dados/PCEPersonalSpending.csv', \n",
    "                          dtype={'Spending':'float'},\n",
    "                         index_col=0, \n",
    "                       parse_dates=True)\n",
    "df = money_df.join(spending_df, on='Date')\n",
    "df.index.freq = 'MS'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "28b6b8cb-0595-4b84-8739-6eb7844b668f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "title = 'M2 Money Stock vs. Personal Consumption Expenditures'\n",
    "ylabel= 'Billions of dollars'\n",
    "xlabel= ''\n",
    "\n",
    "ax = df['Spending'].plot(figsize=(15,8),\n",
    "                         title=title,\n",
    "                         legend=True)\n",
    "ax.autoscale(axis='x',tight=True)\n",
    "ax.set(xlabel=xlabel, ylabel=ylabel)\n",
    "df['Money'].plot(legend=True);"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8fd894e5-802f-433e-ad43-0885f743f23b",
   "metadata": {},
   "source": [
    "### Curiosidade econômica\n",
    "\n",
    "M2 é um agregado monetário que corresponde ao M1 (papel moeda em poder do público + depósitos à vista) somado aos depósitos a prazo (depósitos para investimentos, depósitos de poupança, fundos de aplicação financeira e de renda fixa de curto prazo) + títulos do governo em poder do público.\n",
    "\n",
    "Os agregados monetários vão até M5, e cada nível possui menos liquidez que o anterior "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "c77b7d70-3651-4dac-b8f1-252b92026036",
   "metadata": {},
   "outputs": [],
   "source": [
    "def dickey_fuller(series,title='df'):\n",
    "    '''Teste de Hipótese para estacionariedade'''\n",
    "    print(f'Augmented Dickey Fuller Test for the dataset {title}')\n",
    "    \n",
    "    result = adfuller(series.dropna(),autolag='AIC')\n",
    "    # use help(adfuller) to understand why these labels are chosen\n",
    "    labels = ['ADF test statistics','p-value','#lags','#observations']\n",
    "    \n",
    "    outcome = pd.Series(result[0:4],index=labels)\n",
    "    \n",
    "    for key,val in result[4].items():\n",
    "        outcome[f'critical value ({key})'] = val\n",
    "        \n",
    "    # this will not print the line 'dtype:float64'\n",
    "    print(outcome.to_string())\n",
    "    \n",
    "    if result[1] <= 0.05:\n",
    "        # H_0 é que os dados não são estacionários\n",
    "        print('Strong evidence against the null hypothesis') \n",
    "        print('Reject the null hypothesis')\n",
    "        print('Data is Stationary')\n",
    "    else:\n",
    "        print('Weak evidence against the Null hypothesis')\n",
    "        print('Fail to reject the null hypothesis')\n",
    "        print('Data has a unit root and is non stationary')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "30a41fc9-1982-4177-97e0-200f3bbe9352",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Augmented Dickey Fuller Test for the dataset M2\n",
      "ADF test statistics       4.239022\n",
      "p-value                   1.000000\n",
      "#lags                     4.000000\n",
      "#observations           247.000000\n",
      "critical value (1%)      -3.457105\n",
      "critical value (5%)      -2.873314\n",
      "critical value (10%)     -2.573044\n",
      "Weak evidence against the Null hypothesis\n",
      "Fail to reject the null hypothesis\n",
      "Data has a unit root and is non stationary\n"
     ]
    }
   ],
   "source": [
    "dickey_fuller(df['Money'], 'M2')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "51759b58-f1c3-4f2d-913c-5ad3d1006312",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Augmented Dickey Fuller Test for the dataset Gastos\n",
      "ADF test statistics       0.149796\n",
      "p-value                   0.969301\n",
      "#lags                     3.000000\n",
      "#observations           248.000000\n",
      "critical value (1%)      -3.456996\n",
      "critical value (5%)      -2.873266\n",
      "critical value (10%)     -2.573019\n",
      "Weak evidence against the Null hypothesis\n",
      "Fail to reject the null hypothesis\n",
      "Data has a unit root and is non stationary\n"
     ]
    }
   ],
   "source": [
    "dickey_fuller(df['Spending'], 'Gastos')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "af7d7cc7-0970-4d9b-9749-642f47c90f19",
   "metadata": {},
   "source": [
    "A análise gráfica já indicava isso, mas os testes confirmam: ambas as séries não são estacionárias."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "ff0ec696-a5f9-407f-9594-9e0ca8ff3c3b",
   "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>Money</th>\n",
       "      <th>Spending</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Date</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1995-02-01</th>\n",
       "      <td>-2.5</td>\n",
       "      <td>-0.4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1995-03-01</th>\n",
       "      <td>1.2</td>\n",
       "      <td>34.6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1995-04-01</th>\n",
       "      <td>8.1</td>\n",
       "      <td>4.8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1995-05-01</th>\n",
       "      <td>25.0</td>\n",
       "      <td>42.9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1995-06-01</th>\n",
       "      <td>24.7</td>\n",
       "      <td>44.4</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            Money  Spending\n",
       "Date                       \n",
       "1995-02-01   -2.5      -0.4\n",
       "1995-03-01    1.2      34.6\n",
       "1995-04-01    8.1       4.8\n",
       "1995-05-01   25.0      42.9\n",
       "1995-06-01   24.7      44.4"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_diff1 = df.diff(1).dropna()\n",
    "df_diff1.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "49b6f040-fda8-489a-85d2-af69c1aed0a0",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Augmented Dickey Fuller Test for the dataset M2 diff\n",
      "ADF test statistics      -2.057404\n",
      "p-value                   0.261984\n",
      "#lags                    15.000000\n",
      "#observations           235.000000\n",
      "critical value (1%)      -3.458487\n",
      "critical value (5%)      -2.873919\n",
      "critical value (10%)     -2.573367\n",
      "Weak evidence against the Null hypothesis\n",
      "Fail to reject the null hypothesis\n",
      "Data has a unit root and is non stationary\n"
     ]
    }
   ],
   "source": [
    "dickey_fuller(df_diff1['Money'], 'M2 diff')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "9ed20fef-a4b5-4303-8c0b-45cc6a099935",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Augmented Dickey Fuller Test for the dataset Gastos diff\n",
      "ADF test statistics    -7.226974e+00\n",
      "p-value                 2.041027e-10\n",
      "#lags                   2.000000e+00\n",
      "#observations           2.480000e+02\n",
      "critical value (1%)    -3.456996e+00\n",
      "critical value (5%)    -2.873266e+00\n",
      "critical value (10%)   -2.573019e+00\n",
      "Strong evidence against the null hypothesis\n",
      "Reject the null hypothesis\n",
      "Data is Stationary\n"
     ]
    }
   ],
   "source": [
    "dickey_fuller(df_diff1['Spending'], 'Gastos diff')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "0a10329f-178c-4602-9dc7-7ed1d8e36be3",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Augmented Dickey Fuller Test for the dataset M2 diff\n",
      "ADF test statistics    -7.077471e+00\n",
      "p-value                 4.760675e-10\n",
      "#lags                   1.400000e+01\n",
      "#observations           2.350000e+02\n",
      "critical value (1%)    -3.458487e+00\n",
      "critical value (5%)    -2.873919e+00\n",
      "critical value (10%)   -2.573367e+00\n",
      "Strong evidence against the null hypothesis\n",
      "Reject the null hypothesis\n",
      "Data is Stationary\n"
     ]
    }
   ],
   "source": [
    "dickey_fuller(df['Money'].diff().diff().dropna(), 'M2 diff')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "76cdab3f-8565-45f8-b908-ec226f450d5b",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Augmented Dickey Fuller Test for the dataset Gastos diff\n",
      "ADF test statistics    -8.760145e+00\n",
      "p-value                 2.687900e-14\n",
      "#lags                   8.000000e+00\n",
      "#observations           2.410000e+02\n",
      "critical value (1%)    -3.457779e+00\n",
      "critical value (5%)    -2.873609e+00\n",
      "critical value (10%)   -2.573202e+00\n",
      "Strong evidence against the null hypothesis\n",
      "Reject the null hypothesis\n",
      "Data is Stationary\n"
     ]
    }
   ],
   "source": [
    "dickey_fuller(df['Spending'].diff().diff().dropna(), 'Gastos diff')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c0148d3d-5ffe-40f0-8a3c-29cf4e1ed61d",
   "metadata": {},
   "source": [
    "Com isso, temos os dados finais para a estacionariedade de cada série."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "0ef031a0-7eb5-4204-a968-d85bee6ab24f",
   "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>Money</th>\n",
       "      <th>Spending</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Date</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1995-03-01</th>\n",
       "      <td>3.7</td>\n",
       "      <td>35.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1995-04-01</th>\n",
       "      <td>6.9</td>\n",
       "      <td>-29.8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1995-05-01</th>\n",
       "      <td>16.9</td>\n",
       "      <td>38.1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1995-06-01</th>\n",
       "      <td>-0.3</td>\n",
       "      <td>1.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1995-07-01</th>\n",
       "      <td>-6.2</td>\n",
       "      <td>-51.7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-08-01</th>\n",
       "      <td>-0.7</td>\n",
       "      <td>-8.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-09-01</th>\n",
       "      <td>5.5</td>\n",
       "      <td>-39.8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-10-01</th>\n",
       "      <td>-23.1</td>\n",
       "      <td>24.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-11-01</th>\n",
       "      <td>55.8</td>\n",
       "      <td>10.7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-12-01</th>\n",
       "      <td>-31.2</td>\n",
       "      <td>-15.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>250 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "            Money  Spending\n",
       "Date                       \n",
       "1995-03-01    3.7      35.0\n",
       "1995-04-01    6.9     -29.8\n",
       "1995-05-01   16.9      38.1\n",
       "1995-06-01   -0.3       1.5\n",
       "1995-07-01   -6.2     -51.7\n",
       "...           ...       ...\n",
       "2015-08-01   -0.7      -8.5\n",
       "2015-09-01    5.5     -39.8\n",
       "2015-10-01  -23.1      24.5\n",
       "2015-11-01   55.8      10.7\n",
       "2015-12-01  -31.2     -15.0\n",
       "\n",
       "[250 rows x 2 columns]"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_transformed = df.diff().diff().dropna()\n",
    "df_transformed"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "da1c4859-522a-4771-b4af-78211d14ca38",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(250, 2)"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_transformed.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d6d0feeb-d407-4fb9-95c4-d85acce1d984",
   "metadata": {},
   "source": [
    "## Divisão em treino e teste"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "01bf4cc3-88e8-4886-b70d-a0cf4a666b2e",
   "metadata": {},
   "outputs": [],
   "source": [
    "n_obs = 12\n",
    "train = df_transformed[0:-n_obs]\n",
    "test = df_transformed[-n_obs:]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a7a51b24-7374-4d0c-8d17-037209378550",
   "metadata": {},
   "source": [
    "## Testar qual ordem $p$ é melhor"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "38700180-3cef-46a8-8484-e36f71154a16",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "ARIMA(maxiter=1500, order=(1, 2, 1), scoring_args={}, suppress_warnings=True,\n",
       "      with_intercept=False)"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "auto_arima(df['Money'],maxiter=1500)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "4751d613-7201-4cfc-870b-f8eb52162c96",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "ARIMA(maxiter=1500, order=(1, 1, 2), scoring_args={}, suppress_warnings=True)"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "auto_arima(df['Spending'],maxiter=1500)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d6d7292b-a672-4b9d-95e6-ce88fbadbba0",
   "metadata": {},
   "source": [
    "Ou seja, o melhor seria a ordem 1 para $p$."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "9267bae0-b627-461f-9f88-2577dc19d9a6",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<caption>Statespace Model Results</caption>\n",
       "<tr>\n",
       "  <th>Dep. Variable:</th>   <td>['Money', 'Spending']</td> <th>  No. Observations:  </th>    <td>238</td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Model:</th>                <td>VARMA(1,2)</td>       <th>  Log Likelihood     </th> <td>-2286.305</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th></th>                      <td>+ intercept</td>      <th>  AIC                </th> <td>4606.610</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Date:</th>              <td>Sun, 14 Aug 2022</td>    <th>  BIC                </th> <td>4665.638</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Time:</th>                  <td>03:14:46</td>        <th>  HQIC               </th> <td>4630.399</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Sample:</th>               <td>03-01-1995</td>       <th>                     </th>     <td> </td>    \n",
       "</tr>\n",
       "<tr>\n",
       "  <th></th>                     <td>- 12-01-2014</td>      <th>                     </th>     <td> </td>    \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Covariance Type:</th>          <td>opg</td>          <th>                     </th>     <td> </td>    \n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "  <th>Ljung-Box (L1) (Q):</th>     <td>0.63, 0.00</td> <th>  Jarque-Bera (JB):  </th> <td>546.15, 111.91</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Prob(Q):</th>                <td>0.43, 1.00</td> <th>  Prob(JB):          </th>   <td>0.00, 0.00</td>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Heteroskedasticity (H):</th> <td>5.54, 2.94</td> <th>  Skew:              </th>   <td>1.32, -0.32</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Prob(H) (two-sided):</th>    <td>0.00, 0.00</td> <th>  Kurtosis:          </th>   <td>9.94, 6.30</td>  \n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<caption>Results for equation Money</caption>\n",
       "<tr>\n",
       "         <td></td>           <th>coef</th>     <th>std err</th>      <th>z</th>      <th>P>|z|</th>  <th>[0.025</th>    <th>0.975]</th>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>intercept</th>      <td>   -0.0679</td> <td>    0.944</td> <td>   -0.072</td> <td> 0.943</td> <td>   -1.918</td> <td>    1.782</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>L1.Money</th>       <td>   -1.0853</td> <td>    3.720</td> <td>   -0.292</td> <td> 0.771</td> <td>   -8.377</td> <td>    6.206</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>L1.Spending</th>    <td>    2.2370</td> <td>    6.645</td> <td>    0.337</td> <td> 0.736</td> <td>  -10.787</td> <td>   15.261</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>L1.e(Money)</th>    <td>    0.3154</td> <td>    3.706</td> <td>    0.085</td> <td> 0.932</td> <td>   -6.948</td> <td>    7.579</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>L1.e(Spending)</th> <td>   -2.3655</td> <td>    6.650</td> <td>   -0.356</td> <td> 0.722</td> <td>  -15.399</td> <td>   10.668</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>L2.e(Money)</th>    <td>   -1.3883</td> <td>    3.897</td> <td>   -0.356</td> <td> 0.722</td> <td>   -9.027</td> <td>    6.250</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>L2.e(Spending)</th> <td>    2.0489</td> <td>    5.766</td> <td>    0.355</td> <td> 0.722</td> <td>   -9.252</td> <td>   13.350</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<caption>Results for equation Spending</caption>\n",
       "<tr>\n",
       "         <td></td>           <th>coef</th>     <th>std err</th>      <th>z</th>      <th>P>|z|</th>  <th>[0.025</th>    <th>0.975]</th>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>intercept</th>      <td>    0.0920</td> <td>    0.354</td> <td>    0.260</td> <td> 0.795</td> <td>   -0.602</td> <td>    0.786</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>L1.Money</th>       <td>   -0.3485</td> <td>    2.179</td> <td>   -0.160</td> <td> 0.873</td> <td>   -4.618</td> <td>    3.921</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>L1.Spending</th>    <td>    0.8262</td> <td>    3.754</td> <td>    0.220</td> <td> 0.826</td> <td>   -6.531</td> <td>    8.183</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>L1.e(Money)</th>    <td>    0.4840</td> <td>    2.244</td> <td>    0.216</td> <td> 0.829</td> <td>   -3.914</td> <td>    4.882</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>L1.e(Spending)</th> <td>   -1.7745</td> <td>    3.726</td> <td>   -0.476</td> <td> 0.634</td> <td>   -9.078</td> <td>    5.529</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>L2.e(Money)</th>    <td>   -0.4810</td> <td>    2.361</td> <td>   -0.204</td> <td> 0.839</td> <td>   -5.108</td> <td>    4.146</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>L2.e(Spending)</th> <td>    0.7940</td> <td>    3.209</td> <td>    0.247</td> <td> 0.805</td> <td>   -5.496</td> <td>    7.084</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<caption>Error covariance matrix</caption>\n",
       "<tr>\n",
       "             <td></td>                <th>coef</th>     <th>std err</th>      <th>z</th>      <th>P>|z|</th>  <th>[0.025</th>    <th>0.975]</th>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>sqrt.var.Money</th>          <td>   25.6306</td> <td>    1.044</td> <td>   24.544</td> <td> 0.000</td> <td>   23.584</td> <td>   27.677</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>sqrt.cov.Money.Spending</th> <td>  -10.3463</td> <td>    1.897</td> <td>   -5.454</td> <td> 0.000</td> <td>  -14.065</td> <td>   -6.628</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>sqrt.var.Spending</th>       <td>   33.5976</td> <td>    1.261</td> <td>   26.633</td> <td> 0.000</td> <td>   31.125</td> <td>   36.070</td>\n",
       "</tr>\n",
       "</table><br/><br/>Warnings:<br/>[1] Covariance matrix calculated using the outer product of gradients (complex-step)."
      ],
      "text/plain": [
       "<class 'statsmodels.iolib.summary.Summary'>\n",
       "\"\"\"\n",
       "                             Statespace Model Results                            \n",
       "=================================================================================\n",
       "Dep. Variable:     ['Money', 'Spending']   No. Observations:                  238\n",
       "Model:                        VARMA(1,2)   Log Likelihood               -2286.305\n",
       "                             + intercept   AIC                           4606.610\n",
       "Date:                   Sun, 14 Aug 2022   BIC                           4665.638\n",
       "Time:                           03:14:46   HQIC                          4630.399\n",
       "Sample:                       03-01-1995                                         \n",
       "                            - 12-01-2014                                         \n",
       "Covariance Type:                     opg                                         \n",
       "===================================================================================\n",
       "Ljung-Box (L1) (Q):             0.63, 0.00   Jarque-Bera (JB):       546.15, 111.91\n",
       "Prob(Q):                        0.43, 1.00   Prob(JB):                   0.00, 0.00\n",
       "Heteroskedasticity (H):         5.54, 2.94   Skew:                      1.32, -0.32\n",
       "Prob(H) (two-sided):            0.00, 0.00   Kurtosis:                   9.94, 6.30\n",
       "                            Results for equation Money                            \n",
       "==================================================================================\n",
       "                     coef    std err          z      P>|z|      [0.025      0.975]\n",
       "----------------------------------------------------------------------------------\n",
       "intercept         -0.0679      0.944     -0.072      0.943      -1.918       1.782\n",
       "L1.Money          -1.0853      3.720     -0.292      0.771      -8.377       6.206\n",
       "L1.Spending        2.2370      6.645      0.337      0.736     -10.787      15.261\n",
       "L1.e(Money)        0.3154      3.706      0.085      0.932      -6.948       7.579\n",
       "L1.e(Spending)    -2.3655      6.650     -0.356      0.722     -15.399      10.668\n",
       "L2.e(Money)       -1.3883      3.897     -0.356      0.722      -9.027       6.250\n",
       "L2.e(Spending)     2.0489      5.766      0.355      0.722      -9.252      13.350\n",
       "                          Results for equation Spending                           \n",
       "==================================================================================\n",
       "                     coef    std err          z      P>|z|      [0.025      0.975]\n",
       "----------------------------------------------------------------------------------\n",
       "intercept          0.0920      0.354      0.260      0.795      -0.602       0.786\n",
       "L1.Money          -0.3485      2.179     -0.160      0.873      -4.618       3.921\n",
       "L1.Spending        0.8262      3.754      0.220      0.826      -6.531       8.183\n",
       "L1.e(Money)        0.4840      2.244      0.216      0.829      -3.914       4.882\n",
       "L1.e(Spending)    -1.7745      3.726     -0.476      0.634      -9.078       5.529\n",
       "L2.e(Money)       -0.4810      2.361     -0.204      0.839      -5.108       4.146\n",
       "L2.e(Spending)     0.7940      3.209      0.247      0.805      -5.496       7.084\n",
       "                                  Error covariance matrix                                  \n",
       "===========================================================================================\n",
       "                              coef    std err          z      P>|z|      [0.025      0.975]\n",
       "-------------------------------------------------------------------------------------------\n",
       "sqrt.var.Money             25.6306      1.044     24.544      0.000      23.584      27.677\n",
       "sqrt.cov.Money.Spending   -10.3463      1.897     -5.454      0.000     -14.065      -6.628\n",
       "sqrt.var.Spending          33.5976      1.261     26.633      0.000      31.125      36.070\n",
       "===========================================================================================\n",
       "\n",
       "Warnings:\n",
       "[1] Covariance matrix calculated using the outer product of gradients (complex-step).\n",
       "\"\"\""
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "modelo_varma = VARMAX(train, order=(1, 2), trend='c')\n",
    "results = modelo_varma.fit(maxiter=10_000, disp=False)\n",
    "results.summary()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c4621910-659f-4f49-91ac-c4f8b7f41d80",
   "metadata": {},
   "source": [
    "## Realizar previsões no período de teste"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "bd586950-7ba9-488d-8013-4521a888b686",
   "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>Money</th>\n",
       "      <th>Spending</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2015-01-01</th>\n",
       "      <td>-11.339133</td>\n",
       "      <td>38.944118</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-02-01</th>\n",
       "      <td>-11.101564</td>\n",
       "      <td>-5.103266</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-03-01</th>\n",
       "      <td>0.564258</td>\n",
       "      <td>-0.255383</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-04-01</th>\n",
       "      <td>-1.251545</td>\n",
       "      <td>-0.315606</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-05-01</th>\n",
       "      <td>0.584348</td>\n",
       "      <td>0.267413</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-06-01</th>\n",
       "      <td>-0.103866</td>\n",
       "      <td>0.109302</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-07-01</th>\n",
       "      <td>0.289332</td>\n",
       "      <td>0.218508</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-08-01</th>\n",
       "      <td>0.106902</td>\n",
       "      <td>0.171707</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-09-01</th>\n",
       "      <td>0.200192</td>\n",
       "      <td>0.196615</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-10-01</th>\n",
       "      <td>0.154668</td>\n",
       "      <td>0.184683</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-11-01</th>\n",
       "      <td>0.177382</td>\n",
       "      <td>0.190690</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-12-01</th>\n",
       "      <td>0.166168</td>\n",
       "      <td>0.187737</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                Money   Spending\n",
       "2015-01-01 -11.339133  38.944118\n",
       "2015-02-01 -11.101564  -5.103266\n",
       "2015-03-01   0.564258  -0.255383\n",
       "2015-04-01  -1.251545  -0.315606\n",
       "2015-05-01   0.584348   0.267413\n",
       "2015-06-01  -0.103866   0.109302\n",
       "2015-07-01   0.289332   0.218508\n",
       "2015-08-01   0.106902   0.171707\n",
       "2015-09-01   0.200192   0.196615\n",
       "2015-10-01   0.154668   0.184683\n",
       "2015-11-01   0.177382   0.190690\n",
       "2015-12-01   0.166168   0.187737"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_forecast = results.forecast(n_obs)\n",
    "df_forecast"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6042cf36-92f4-424d-98c5-58696a66071e",
   "metadata": {},
   "source": [
    "## Desfazer as transformações"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "6cbd863a-cfa5-4f25-a5f5-940017be2160",
   "metadata": {},
   "outputs": [],
   "source": [
    "df_forecast['Money1d'] = (df['Money'].iloc[-n_obs-1]-df['Money'].iloc[-n_obs-2]) + df_forecast['Money'].cumsum()\n",
    "df_forecast['MoneyForecast'] = df['Money'].iloc[-n_obs-1] + df_forecast['Money'].cumsum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "f5e0e51c-84f5-45d9-abd8-e013389d98e4",
   "metadata": {},
   "outputs": [],
   "source": [
    "df_forecast['Spending1d'] = (df['Spending'].iloc[-n_obs-1]-df['Spending'].iloc[-n_obs-2]) + df_forecast['Spending'].cumsum()\n",
    "df_forecast['SpendingForecast'] = df['Spending'].iloc[-n_obs-1] + df_forecast['Spending'].cumsum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "875c0cb8-0ee1-42f4-80ce-7b365ac3f20d",
   "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>Money</th>\n",
       "      <th>Spending</th>\n",
       "      <th>Money1d</th>\n",
       "      <th>MoneyForecast</th>\n",
       "      <th>Spending1d</th>\n",
       "      <th>SpendingForecast</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2015-01-01</th>\n",
       "      <td>-11.339133</td>\n",
       "      <td>38.944118</td>\n",
       "      <td>67.260867</td>\n",
       "      <td>11658.760867</td>\n",
       "      <td>49.544118</td>\n",
       "      <td>12100.944118</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-02-01</th>\n",
       "      <td>-11.101564</td>\n",
       "      <td>-5.103266</td>\n",
       "      <td>56.159303</td>\n",
       "      <td>11647.659303</td>\n",
       "      <td>44.440852</td>\n",
       "      <td>12095.840852</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-03-01</th>\n",
       "      <td>0.564258</td>\n",
       "      <td>-0.255383</td>\n",
       "      <td>56.723562</td>\n",
       "      <td>11648.223562</td>\n",
       "      <td>44.185469</td>\n",
       "      <td>12095.585469</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-04-01</th>\n",
       "      <td>-1.251545</td>\n",
       "      <td>-0.315606</td>\n",
       "      <td>55.472016</td>\n",
       "      <td>11646.972016</td>\n",
       "      <td>43.869863</td>\n",
       "      <td>12095.269863</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-05-01</th>\n",
       "      <td>0.584348</td>\n",
       "      <td>0.267413</td>\n",
       "      <td>56.056364</td>\n",
       "      <td>11647.556364</td>\n",
       "      <td>44.137276</td>\n",
       "      <td>12095.537276</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-06-01</th>\n",
       "      <td>-0.103866</td>\n",
       "      <td>0.109302</td>\n",
       "      <td>55.952498</td>\n",
       "      <td>11647.452498</td>\n",
       "      <td>44.246578</td>\n",
       "      <td>12095.646578</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-07-01</th>\n",
       "      <td>0.289332</td>\n",
       "      <td>0.218508</td>\n",
       "      <td>56.241830</td>\n",
       "      <td>11647.741830</td>\n",
       "      <td>44.465086</td>\n",
       "      <td>12095.865086</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-08-01</th>\n",
       "      <td>0.106902</td>\n",
       "      <td>0.171707</td>\n",
       "      <td>56.348732</td>\n",
       "      <td>11647.848732</td>\n",
       "      <td>44.636793</td>\n",
       "      <td>12096.036793</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-09-01</th>\n",
       "      <td>0.200192</td>\n",
       "      <td>0.196615</td>\n",
       "      <td>56.548923</td>\n",
       "      <td>11648.048923</td>\n",
       "      <td>44.833408</td>\n",
       "      <td>12096.233408</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-10-01</th>\n",
       "      <td>0.154668</td>\n",
       "      <td>0.184683</td>\n",
       "      <td>56.703591</td>\n",
       "      <td>11648.203591</td>\n",
       "      <td>45.018091</td>\n",
       "      <td>12096.418091</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-11-01</th>\n",
       "      <td>0.177382</td>\n",
       "      <td>0.190690</td>\n",
       "      <td>56.880973</td>\n",
       "      <td>11648.380973</td>\n",
       "      <td>45.208781</td>\n",
       "      <td>12096.608781</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-12-01</th>\n",
       "      <td>0.166168</td>\n",
       "      <td>0.187737</td>\n",
       "      <td>57.047141</td>\n",
       "      <td>11648.547141</td>\n",
       "      <td>45.396518</td>\n",
       "      <td>12096.796518</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                Money   Spending    Money1d  MoneyForecast  Spending1d  \\\n",
       "2015-01-01 -11.339133  38.944118  67.260867   11658.760867   49.544118   \n",
       "2015-02-01 -11.101564  -5.103266  56.159303   11647.659303   44.440852   \n",
       "2015-03-01   0.564258  -0.255383  56.723562   11648.223562   44.185469   \n",
       "2015-04-01  -1.251545  -0.315606  55.472016   11646.972016   43.869863   \n",
       "2015-05-01   0.584348   0.267413  56.056364   11647.556364   44.137276   \n",
       "2015-06-01  -0.103866   0.109302  55.952498   11647.452498   44.246578   \n",
       "2015-07-01   0.289332   0.218508  56.241830   11647.741830   44.465086   \n",
       "2015-08-01   0.106902   0.171707  56.348732   11647.848732   44.636793   \n",
       "2015-09-01   0.200192   0.196615  56.548923   11648.048923   44.833408   \n",
       "2015-10-01   0.154668   0.184683  56.703591   11648.203591   45.018091   \n",
       "2015-11-01   0.177382   0.190690  56.880973   11648.380973   45.208781   \n",
       "2015-12-01   0.166168   0.187737  57.047141   11648.547141   45.396518   \n",
       "\n",
       "            SpendingForecast  \n",
       "2015-01-01      12100.944118  \n",
       "2015-02-01      12095.840852  \n",
       "2015-03-01      12095.585469  \n",
       "2015-04-01      12095.269863  \n",
       "2015-05-01      12095.537276  \n",
       "2015-06-01      12095.646578  \n",
       "2015-07-01      12095.865086  \n",
       "2015-08-01      12096.036793  \n",
       "2015-09-01      12096.233408  \n",
       "2015-10-01      12096.418091  \n",
       "2015-11-01      12096.608781  \n",
       "2015-12-01      12096.796518  "
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_forecast"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "c4e6a9cf-7216-4444-89d1-ec2a8fdce390",
   "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>Money</th>\n",
       "      <th>Spending</th>\n",
       "      <th>MoneyForecast</th>\n",
       "      <th>SpendingForecast</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2015-01-01</th>\n",
       "      <td>11733.2</td>\n",
       "      <td>12046.0</td>\n",
       "      <td>11658.760867</td>\n",
       "      <td>12100.944118</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-02-01</th>\n",
       "      <td>11852.4</td>\n",
       "      <td>12082.4</td>\n",
       "      <td>11647.659303</td>\n",
       "      <td>12095.840852</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-03-01</th>\n",
       "      <td>11868.8</td>\n",
       "      <td>12158.3</td>\n",
       "      <td>11648.223562</td>\n",
       "      <td>12095.585469</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-04-01</th>\n",
       "      <td>11916.1</td>\n",
       "      <td>12193.8</td>\n",
       "      <td>11646.972016</td>\n",
       "      <td>12095.269863</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-05-01</th>\n",
       "      <td>11947.6</td>\n",
       "      <td>12268.1</td>\n",
       "      <td>11647.556364</td>\n",
       "      <td>12095.537276</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-06-01</th>\n",
       "      <td>11993.1</td>\n",
       "      <td>12308.3</td>\n",
       "      <td>11647.452498</td>\n",
       "      <td>12095.646578</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-07-01</th>\n",
       "      <td>12045.3</td>\n",
       "      <td>12355.4</td>\n",
       "      <td>11647.741830</td>\n",
       "      <td>12095.865086</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-08-01</th>\n",
       "      <td>12096.8</td>\n",
       "      <td>12394.0</td>\n",
       "      <td>11647.848732</td>\n",
       "      <td>12096.036793</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-09-01</th>\n",
       "      <td>12153.8</td>\n",
       "      <td>12392.8</td>\n",
       "      <td>11648.048923</td>\n",
       "      <td>12096.233408</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-10-01</th>\n",
       "      <td>12187.7</td>\n",
       "      <td>12416.1</td>\n",
       "      <td>11648.203591</td>\n",
       "      <td>12096.418091</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-11-01</th>\n",
       "      <td>12277.4</td>\n",
       "      <td>12450.1</td>\n",
       "      <td>11648.380973</td>\n",
       "      <td>12096.608781</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-12-01</th>\n",
       "      <td>12335.9</td>\n",
       "      <td>12469.1</td>\n",
       "      <td>11648.547141</td>\n",
       "      <td>12096.796518</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              Money  Spending  MoneyForecast  SpendingForecast\n",
       "2015-01-01  11733.2   12046.0   11658.760867      12100.944118\n",
       "2015-02-01  11852.4   12082.4   11647.659303      12095.840852\n",
       "2015-03-01  11868.8   12158.3   11648.223562      12095.585469\n",
       "2015-04-01  11916.1   12193.8   11646.972016      12095.269863\n",
       "2015-05-01  11947.6   12268.1   11647.556364      12095.537276\n",
       "2015-06-01  11993.1   12308.3   11647.452498      12095.646578\n",
       "2015-07-01  12045.3   12355.4   11647.741830      12095.865086\n",
       "2015-08-01  12096.8   12394.0   11647.848732      12096.036793\n",
       "2015-09-01  12153.8   12392.8   11648.048923      12096.233408\n",
       "2015-10-01  12187.7   12416.1   11648.203591      12096.418091\n",
       "2015-11-01  12277.4   12450.1   11648.380973      12096.608781\n",
       "2015-12-01  12335.9   12469.1   11648.547141      12096.796518"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.concat([df.iloc[-12:],df_forecast[['MoneyForecast','SpendingForecast']]],axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "dca8a8b2-37b2-4b87-9f66-622929563c86",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "df['Money'][-n_obs:].plot(figsize=(12,5),legend=True).autoscale(axis='x',tight=True)\n",
    "df_forecast['MoneyForecast'].plot(legend=True);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "74c52ffe-5bc0-4c13-8a35-d022fbf8b5c7",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "df['Spending'][-n_obs:].plot(figsize=(12,5),legend=True).autoscale(axis='x',tight=True)\n",
    "df_forecast['SpendingForecast'].plot(legend=True);"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5677a053-71d5-440d-b9fb-1b67f5cce722",
   "metadata": {},
   "source": [
    "## Avaliação do Modelo VARMA"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "9ec41cc8-5005-4a28-ac1a-fde569a73e54",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Money VAR(5) RMSE: 423.538\n"
     ]
    }
   ],
   "source": [
    "RMSE1 = rmse(df['Money'][-n_obs:], df_forecast['MoneyForecast'])\n",
    "print(f'Money VAR(5) RMSE: {RMSE1:.3f}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "aeb535be-b86f-4d9c-84ba-216e51bef11e",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Spending VAR(5) RMSE: 241.815\n"
     ]
    }
   ],
   "source": [
    "RMSE2 = rmse(df['Spending'][-n_obs:], df_forecast['SpendingForecast'])\n",
    "print(f'Spending VAR(5) RMSE: {RMSE2:.3f}')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "159e8df9-3965-4112-aad5-d1e178b57dc1",
   "metadata": {},
   "source": [
    "O modelo VARMA teve péssima avaliação nestes dados. Talvez a cointegração não fosse boa ou as quebras do período (crises de 98/99, bolha.com e de 2008) interfiram nisso também."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e3ca56b3-797d-46ba-aa7b-d46d58d54aef",
   "metadata": {},
   "source": [
    "## Análise de modelo VAR"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "cffc8f10-6c01-4b70-a72d-eb5a98122564",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "VAR Order 1\n",
      "AIC 14.178610495220896 \n",
      "\n",
      "VAR Order 2\n",
      "AIC 13.955189367163705 \n",
      "\n",
      "VAR Order 3\n",
      "AIC 13.849518291541038 \n",
      "\n",
      "VAR Order 4\n",
      "AIC 13.827950574458283 \n",
      "\n",
      "VAR Order 5\n",
      "AIC 13.78730034460964 \n",
      "\n",
      "VAR Order 6\n",
      "AIC 13.799076756885809 \n",
      "\n",
      "VAR Order 7\n",
      "AIC 13.797638727913972 \n",
      "\n"
     ]
    }
   ],
   "source": [
    "p = [i for i in range(1, 8)]\n",
    "\n",
    "for i in p:\n",
    "    model = VAR(train)\n",
    "    results = model.fit(i)\n",
    "    print(f'VAR Order {i}')\n",
    "    print('AIC {}'.format(results.aic), '\\n')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "fcc5e9fd-62e6-482d-8ae3-a7fe64910a08",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "  Summary of Regression Results   \n",
       "==================================\n",
       "Model:                         VAR\n",
       "Method:                        OLS\n",
       "Date:           Sun, 14, Aug, 2022\n",
       "Time:                     12:30:54\n",
       "--------------------------------------------------------------------\n",
       "No. of Equations:         2.00000    BIC:                    14.1131\n",
       "Nobs:                     233.000    HQIC:                   13.9187\n",
       "Log likelihood:          -2245.45    FPE:                    972321.\n",
       "AIC:                      13.7873    Det(Omega_mle):         886628.\n",
       "--------------------------------------------------------------------\n",
       "Results for equation Money\n",
       "==============================================================================\n",
       "                 coefficient       std. error           t-stat            prob\n",
       "------------------------------------------------------------------------------\n",
       "const               0.516683         1.782238            0.290           0.772\n",
       "L1.Money           -0.646232         0.068177           -9.479           0.000\n",
       "L1.Spending        -0.107411         0.051388           -2.090           0.037\n",
       "L2.Money           -0.497482         0.077749           -6.399           0.000\n",
       "L2.Spending        -0.192202         0.068613           -2.801           0.005\n",
       "L3.Money           -0.234442         0.081004           -2.894           0.004\n",
       "L3.Spending        -0.178099         0.074288           -2.397           0.017\n",
       "L4.Money           -0.295531         0.075294           -3.925           0.000\n",
       "L4.Spending        -0.035564         0.069664           -0.511           0.610\n",
       "L5.Money           -0.162399         0.066700           -2.435           0.015\n",
       "L5.Spending        -0.058449         0.051357           -1.138           0.255\n",
       "==============================================================================\n",
       "\n",
       "Results for equation Spending\n",
       "==============================================================================\n",
       "                 coefficient       std. error           t-stat            prob\n",
       "------------------------------------------------------------------------------\n",
       "const               0.203469         2.355446            0.086           0.931\n",
       "L1.Money            0.188105         0.090104            2.088           0.037\n",
       "L1.Spending        -0.878970         0.067916          -12.942           0.000\n",
       "L2.Money            0.053017         0.102755            0.516           0.606\n",
       "L2.Spending        -0.625313         0.090681           -6.896           0.000\n",
       "L3.Money           -0.022172         0.107057           -0.207           0.836\n",
       "L3.Spending        -0.389041         0.098180           -3.963           0.000\n",
       "L4.Money           -0.170456         0.099510           -1.713           0.087\n",
       "L4.Spending        -0.245435         0.092069           -2.666           0.008\n",
       "L5.Money           -0.083165         0.088153           -0.943           0.345\n",
       "L5.Spending        -0.181699         0.067874           -2.677           0.007\n",
       "==============================================================================\n",
       "\n",
       "Correlation matrix of residuals\n",
       "               Money  Spending\n",
       "Money       1.000000 -0.267934\n",
       "Spending   -0.267934  1.000000\n",
       "\n"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "results = model.fit(5)\n",
    "results.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "cbb09b34-817c-4ab5-a308-6e06cdb3624f",
   "metadata": {},
   "outputs": [],
   "source": [
    "z = results.forecast(y=train.values[-5:],steps = n_obs)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "8071d382-e437-4e11-a641-348470a531ed",
   "metadata": {},
   "outputs": [],
   "source": [
    "idx = pd.date_range(start='1/1/2015', periods=12, freq='MS')\n",
    "df_forecast = pd.DataFrame(z,index=idx,columns=['Money2D','Spending2D'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "1adbfced-a13f-4629-b9a7-b05c37ef1fe2",
   "metadata": {},
   "outputs": [],
   "source": [
    "df_forecast['Money1D'] = (df['Money'].iloc[-n_obs-1] - df['Money'].iloc[-n_obs-2]) + df_forecast['Money2D'].cumsum()\n",
    "df_forecast['MoneyForecast'] = df['Money'].iloc[-n_obs-1] + df_forecast['Money1D'].cumsum()\n",
    "\n",
    "df_forecast['Spending1D'] = (df['Spending'].iloc[-n_obs-1]-df['Spending'].iloc[-n_obs-2]) + df_forecast['Spending2D'].cumsum()\n",
    "df_forecast['SpendingForecast'] = df['Spending'].iloc[-n_obs-1] + df_forecast['Spending1D'].cumsum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "id": "6ea7efb6-3f53-45d4-8bbd-5574a22d2d9f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAtwAAAFICAYAAABjtimhAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/YYfK9AAAACXBIWXMAAAsTAAALEwEAmpwYAABfkElEQVR4nO3dd3yW9fX/8dfJhpCEkYRAwgojjEACBMQ6QJClKLhRq1bbau2ytna3ane1/XWo/fZb+3VUBcStVREH4MYWJEDC3iQBEhISAtm5P78/rpsQkU2SO3fyfj4eeZD7WjkXIcnhk3OdY845RERERESkeYQEOgARERERkbZMCbeIiIiISDNSwi0iIiIi0oyUcIuIiIiINCMl3CIiIiIizSgs0AGcrvj4eNe3b99AhyEiIiIibdzy5cv3OucSTvf8oE24+/bty7JlywIdhoiIiIi0cWa2/UzOV0mJiIiIiEgzUsItIiIiItKMTphwm9mjZlZoZjmNtv3BzNaZ2Soze9HMOh9xTm8zO2BmdzXaNtrMVpvZJjN7wMzMvz3SzOb7t39iZn2b7vZERERERALrZGq4HwceAp5otO0t4MfOuTozuw/4MfDDRvv/DCw44jp/B24FlgKvA9P8x3wZ2OecG2Bms4H7gGtO/VagtraWvLw8qqqqTud0CZCoqChSUlIIDw8PdCgiIiIiTe6ECbdz7r0jV52dc282erkUuPLQCzObBWwBDjba1gOIdc597H/9BDALL+GeCdzrP/Q54CEzM+ecO9WbycvLIyYmhr59++JfQJdWzjlHcXExeXl59OvXL9DhiIiIiDS5pqjhvgX/araZReOtdP/iiGOSgbxGr/P82w7t2wngnKsDyoBupxNIVVUV3bp1U7IdRMyMbt266bcSIiIi0madUcJtZj8F6oA5/k2/AP7snDtw5KFHOd2dxL4jP96tZrbMzJYVFRUdK6YTxi2tiz5nIiIi0paddh9uM7sJmAFMalT+cRZwpZndD3QGfGZWBTwPpDQ6PQUo8L+fB/QC8swsDIgDSo72MZ1zDwMPA2RlZZ1yyYmIiIiISEs7rRVuM5uGVzpyqXOu4tB259x5zrm+zrm+wF+A3zrnHnLO7QLKzWycvzvJjcDL/tNeAW7yv38lsOh06rdbCzPjhhtuaHhdV1dHQkICM2bMCGBUIiIiIhIoJ9MWcB7wMZBmZnlm9mW8riUxwFtmlm1m/3sSH+t24P+ATcBmDncxeQToZmabgO8CPzr122g9oqOjycnJobKyEoC33nqL5OTkE5wlIiIiIq1STcWJjzmBEybczrlrnXM9nHPhzrkU59wjzrkBzrlezrlM/9vXjnLevc65PzZ6vcw5l+6c6++c++ahVWznXJVz7ir/Ncc657ac8V0F2PTp03nttdcAmDdvHtdee23DvpKSEmbNmsWIESMYN24cq1atAuDee+/llltuYcKECaSmpvLAAw80nPPUU08xduxYMjMzue2226ivr+eRRx7hzjvvbDjmn//8J9/97ndb6A5FRERE2jDnYMdSeOVb8MdBZ3y5067hbu1+8e9c1hTsb9JrDu0Zyz2XDDvhcbNnz+aXv/wlM2bMYNWqVdxyyy28//77ANxzzz2MHDmSl156iUWLFnHjjTeSnZ0NwLp161i8eDHl5eWkpaVx++23s2nTJubPn8+HH35IeHg4X//615kzZw6zZ89mxIgR3H///YSHh/PYY4/xj3/8o0nvV0RERKRdKcuDlU9D9lwo2Qzh0TBsFt44mdPXZhPuQBoxYgTbtm1j3rx5XHTRRZ/Z98EHH/D8888DMHHiRIqLiykrKwPg4osvJjIyksjISBITE9mzZw/vvPMOy5cvZ8yYMQBUVlaSmJhIdHQ0EydO5NVXX2XIkCHU1tYyfPjwlr1RERERkWBXWwnrXoPsObB5MeCgz7lw3vdg6EyI7IQS7mM4mZXo5nTppZdy1113sWTJEoqLixu2H+150ENt8SIjIxu2hYaGUldXh3OOm266id/97nefO+8rX/kKv/3tbxk8eDA333xzM9yFiIiISBvkHOQvhxVPQc4LUF0Gcb1h/A8gYzZ0TW3SD9dmE+5Au+WWW4iLi2P48OEsWbKkYfv555/PnDlz+PnPf86SJUuIj48nNjb2mNeZNGkSM2fO5M477yQxMZGSkhLKy8vp06cPZ511Fjt37uTTTz9tqAUXERERkWPYvwtWzfdKRvauh7AO3ip25nXQ9zwIaYqZkJ+nhLuZpKSkcMcdd3xu+7333svNN9/MiBEj6NixI//617+Oe52hQ4fy61//milTpuDz+QgPD+dvf/sbffr0AeDqq68mOzubLl26NMt9iIiIiAS1umpY/7qXZG96G5wPeo2DSx+EobMg6tgLn03FgrXldVZWllu2bNlntq1du5YhQ4YEKKLAmDFjBnfeeSeTJk0KdChnpD1+7kRERKSZOAcFK7wke/WzUFUKscleuUjm9dCt/yldzsyWO+eyTjccrXAHqdLSUsaOHUtGRkbQJ9siIiIiTeJA4eGSkcI1EBYFg2d4JSOpEyAkNCBhKeEOUp07d2bDhg2BDkNEREQksOpqYONCWDEHNr4Jrh5SxsCMP8Owy6FD50BHqIRbRERERILQrlX+kpFnoKIYOiXBF77lrWYnpAU6us9Qwi0iIiIiweHgXq8mO3sO7F4NoREw+GKvLjv1Aghtnalt64xKRERERASgvhY2vuUl2RsWgq8Weo6Ei/4I6VdAx66BjvCElHCLiIiISOuzZ42XZK+aDweLIDoBzrrNW83uPjTQ0Z2S5unu3Y6ZGTfccEPD67q6OhISEpgxY0aLxRAaGkpmZmbD27Zt21rsYx/PX/7yFyoqKgIdhoiIiLRWFSXwn3/CP8bD38+GT/4BvcfBtU/Dd9fC1N8EXbINWuFuctHR0eTk5FBZWUmHDh146623SE5ObtEYOnToQHZ29imfV1dXR1hY8/2T+Mtf/sIXv/hFOnbs2GwfQ0RERIJMfR1sXuStZq9/HeprIGkETLsPhl8F0d0CHeEZ0wp3M5g+fTqvvfYaAPPmzePaa69t2FdSUsKsWbMYMWIE48aNaxjJfu+993LLLbcwYcIEUlNTeeCBBxrOeeqppxg7diyZmZncdttt1NfX88gjj3DnnXc2HPPPf/6T7373u8eMKTs7m3HjxjFixAguu+wy9u3bB8CECRP4yU9+wvjx4/nrX//K8uXLGT9+PKNHj2bq1Kns2rULgE2bNnHhhReSkZHBqFGj2Lx5MwcOHGDSpEmMGjWK4cOH8/LLLwNw8OBBLr74YjIyMkhPT2f+/Pk88MADFBQUcMEFF3DBBRc00d+0iIiIBK2i9fDW3fDnYTD3Ktj2Poz5CnztA/ja+zDua20i2Ya2vMK94Efe06tNKWk4TP/9CQ+bPXs2v/zlL5kxYwarVq3illtu4f333wfgnnvuYeTIkbz00kssWrSIG2+8sWE1et26dSxevJjy8nLS0tK4/fbb2bRpE/Pnz+fDDz8kPDycr3/968yZM4fZs2czYsQI7r//fsLDw3nsscf4xz/+AUBlZSWZmZkA9OvXjxdffJEbb7yRBx98kPHjx3P33Xfzi1/8gr/85S+AN0Tn3Xffpba2lvHjx/Pyyy+TkJDA/Pnz+elPf8qjjz7K9ddfz49+9CMuu+wyqqqq8Pl8RERE8OKLLxIbG8vevXsZN24cl156KW+88QY9e/Zs+E9HWVkZcXFx/OlPf2Lx4sXEx8c37edFREREgkNlKeQ877Xzy18GFgqDpnp12QOnQFhEoCNsFm034Q6gESNGsG3bNubNm8dFF130mX0ffPABzz//PAATJ06kuLiYsrIyAC6++GIiIyOJjIwkMTGRPXv28M4777B8+XLGjBkDeMl0YmIi0dHRTJw4kVdffZUhQ4ZQW1vL8OHDgc+XlJSVlVFaWsr48eMBuOmmm7jqqqsa9l9zzTUArF+/npycHCZPngxAfX09PXr0oLy8nPz8fC677DIAoqKiAKitreUnP/kJ7733HiEhIeTn57Nnzx6GDx/OXXfdxQ9/+ENmzJjBeeed16R/vyIiIhJEfPWwZbGXZK99FeqrIXEoTPkNjLgaOiUGOsJm13YT7pNYiW5Ol156KXfddRdLliyhuLi4Ybtz7nPHmhkAkZGRDdtCQ0Opq6vDOcdNN93E7373u8+d95WvfIXf/va3DB48mJtvvvm0Y42Ojm6IbdiwYXz88cef2b9///6jnjdnzhyKiopYvnw54eHh9O3bl6qqKgYNGsTy5ct5/fXX+fGPf8yUKVO4++67Tzs+ERERCUJ7N3l12SufhvIC6NAFRt/kDabpkQn+/Ke1219Ve8bXUA13M7nlllu4++67G1adDzn//POZM2cOAEuWLCE+Pp7Y2NhjXmfSpEk899xzFBYWAl4N+Pbt2wE466yz2LlzJ3Pnzv1MnfiR4uLi6NKlS0NZy5NPPtmw2t1YWloaRUVFDQl3bW0tubm5xMbGkpKSwksvvQRAdXU1FRUVlJWVkZiYSHh4OIsXL26Iq6CggI4dO/LFL36Ru+66i08//RSAmJgYysvLT/h3JyIiIkGqaj8s/xc8MgUeGg0f/sUryb3qX/C99XDRH7we2q082a6t9/HO2j18Y+6njPn122d8vba7wh1gKSkp3HHHHZ/bfu+993LzzTczYsQIOnbsyL/+9a/jXmfo0KH8+te/ZsqUKfh8PsLDw/nb3/5Gnz59ALj66qvJzs6mS5cux73Ov/71L772ta9RUVFBamoqjz322OeOiYiI4LnnnuPb3/42ZWVl1NXV8Z3vfIdhw4bx5JNPctttt3H33XcTHh7Os88+y/XXX88ll1xCVlYWmZmZDB48GIDVq1fz/e9/n5CQEMLDw/n73/8OwK233sr06dPp0aMHixcvPqm/RxEREWnlfD7Y9p5XMrLmFairhPg0mPxLGHENxCQFOsKT4pxjVV4ZL67I598rCyg+WEPX6Ahmj+nFL8/w2na0EodgkJWV5ZYtW/aZbWvXrmXIkCEBiigwZsyYwZ133smkSZMCHcoZaY+fOxERkaBWsgWy58HKeVC2EyLjYPgVkPlFSB7V6lexD9lZUsHL2fm8sCKfLUUHiQgLYfKQ7lw2MpnxaQmEh4ZgZsudc1mn+zG0wh2kSktLGTt2LBkZGUGfbIuIiEiQqD4Aa17yVrO3fwgY9J8Ik38BaRdDeFSgIzwpZZW1vL56Fy9+ms9/tpUAcFa/rtx6XirTh/cgrkN4k348JdxBqnPnzmzYsCHQYYiIiEhb55yXXGfPhdyXoPYgdBsAk+6GEbMhrmUH/J2umjof724o4sUVeby9tpCaOh/9E6L5/tQ0Zmb2JKVL8w3ma3MJt3OuoeuHBIdgLWsSERFp00p3euUi2XNg3zaIiDlcMtJrbFCUjDjnyN5Z2lCXva+ilm7REVw3tjeXj0pmeHJci+SNbSrhjoqKori4mG7duinpDhLOOYqLixt6e4uIiEgA1VTAuldhxVOw9T3AQb/zYcJPYMglENF8q8BNaUdxBS+uyOel7Hy27j1IZFgIk4d25/JRyZw30KvLbkltKuFOSUkhLy+PoqKiQIcipyAqKoqUlJRAhyEiItI+OQd5yyD7Kch5Aar3Q+feMOFHkHEtdOkT6AhPSllFLa+uLuDFT/NZtn0fZjCuXzdun9CfaelJxEY1bV32qWhTCXd4eDj9+vULdBgiIiIird/+XbDqaa82e+8GCO8IQ2d6Y9b7nAMhrX9cS3VdPYvXFfHSinwWrSukpt7HwMRO/GBaGrMyk+nZuUOgQwTaWMItIiIiIsdRVw3rF3h12ZveBueD3mfDpQ/BsFkQGRPoCE/IOcenO/bxwqf5vLZ6F6UVtcR3iuSL4/pw+ahkhvWMbXWlxUq4RURERNoy52DXSi/JXv0sVO6D2GQ4905vNbtb/0BHeFK27T3YUJe9vbiCqPAQpgxN4rJRyZw3IJ6wFq7LPhUnTLjN7FFgBlDonEv3b/sDcAlQA2wGbnbOlZrZWODhQ6cC9zrnXvSfMxp4HOgAvA7c4ZxzZhYJPAGMBoqBa5xz25rsDkVERETao4N7YdUzXqK9JwdCI2HIDC/JTp0AIaGBjvCE9h2s4dXVu3jx0zw+3VGKGXyhfze+NXEg09KT6BQZHGvHJ5w0aWbnAweAJxol3FOARc65OjO7D8A590Mz6wjU+Lf3AFYCPf2v/wPcASzFS7gfcM4tMLOvAyOcc18zs9nAZc65a04U+NEmTYqIiIi0a/W1sPEtL8ne8Ab46iB5NGReB+lXQIcugY7whKrr6lm0tpAXVuSzZH0htfWOtO4xXDYqmZmZPekR1/J12c0+adI5956Z9T1i25uNXi4FrvRvr2i0PQpw/iB7ALHOuY/9r58AZgELgJnAvf5zngMeMjNzas4sIiIicnL2rPGS7FXz4WARRCfCuNu91ezEIYGO7oSccyzb7q/LXlXA/qo6EmIiuensvlw2KpmhPVpfXfapaIp1+FuA+YdemNlZwKNAH+AG/+p2MpDX6Jw84NBYomRgJ4D/2DKgG7D3yA9kZrcCtwL07t27CUIXERERCVKV+2D1c16iXbACQsIhbZqXZA+4EEID1wbvZG0pOsBLK/J5MTufnSWVdAgPZVp6EpeNTOacAfGEhgRvkt3YGSXcZvZToA6Yc2ibc+4TYJiZDQH+ZWYL8Oq5j3RoBft4+z670bmH8deIZ2VlaQVcRERE2hdfPWxe7CXZ616D+mroPhym/R6GXwXR8YGO8IRKDtbw75UFvLAin5U7SwkxOGdAPHdeOIipw5KIDpK67FNx2ndkZjfhPUw56WjlH865tWZ2EEjHW9FuPNkkBSjwv58H9ALyzCwMiANKTjcuERERkTZn7yYvyV75NJQXQIeukHWzV5vdIyPQ0Z1QVW0976wt5MUVeSxZX0SdzzE4KYafXDSYmZnJdI9t2xOnTyvhNrNpwA+B8Y3rts2sH7DTXxrSB0gDtjnn9ppZuZmNAz4BbgQe9J/2CnAT8DFeLfgi1W+LiIhIu1e1H3Jf9AbT7FwKFgIDJsP038OgaRAWGegIj8vnc/x3WwkvrvD6ZZdX1dE9NpIvn9uPWSOTGdIjNtAhtpiTaQs4D5gAxJtZHnAP8GMgEnjLX8C+1Dn3NeBc4EdmVgv4gK875w7VYt/O4baAC/xvAI8AT5rZJryV7dlNcmciIiIiwcbng+0fwIo5sPYVqK2A+EFw4S8gYzbEJAU6whPaVHiAF1fk8dKKAvJLK+kY4dVlXz4yhbP7d2szddmn4oRtAVsrtQUUERGRNmPfdm8le+VcKN0BkXGQfjmM/KLX1q+Vd+jYe6Caf68s4MUV+azKKyPE4NyBCVw+Mpkpw7rTMSK467KbvS2giIiIiDSDmgpvFXvFU7DtfcC8gTST7oHBF0N4y/ebPhVVtfW8tWYPL67I590NRdT7HEN7xPKzi4dwaUZPEtt4XfapUMItIiIi0lKcg52feEl27ktQUw5d+sEFP/NKRjr3CnSEx+XzOZZuLebFT/NZkLObA9V19IiL4qvnpXLZyGTSkmICHWKrpIRbREREpLmV5cPKeV7ZSMlmCI+GYZfByOuh99mtvmTEOcfrq3dz/8J1bC+uIDoilOnDe3D5yGTOSm2fddmnQgm3iIiISHOorYL1r3kPQG5ZDM4Hfc6F874HQ2dCZKdAR3hSlm/fx29eW8OnO0oZnBTDX2dnMmVoEh0iQgMdWtBQwi0iIiLSVJyDgk+9JDvnOagqg7hecN5dkHktdE0NdIQnbXvxQe5/Yz2vrd5FYkwk918xgitGp2g1+zQo4RYRERE5UwcKYdV8L9EuWgthUTDkUm8wTb/xEBIS6AhPWmlFDQ8u2sQTH28jLCSE71w4kK+el9omJ0C2FP3NiYiIiJyOuhrYuNCry96wEFw9pIyBGX/xWvpFxQU6wlNSU+fjiY+38eCiTeyvquXq0b347pRBbX4KZEtQwi0iIiJyspyDXdneiPXVz0JFMXRKgi98y1vNTkgLdISnzDnHgpzd3PeG90DkeQPj+clFQ9rVJMjmpoRbRERE5ET274LVz0D2PK9kJDQSBl8EGddB/4kQGpwp1ac79vGb19ayfPs+BnXvxOM3j2FCWmKgw2pzgvNfh4iIiEhzq6mA9a97JSOHuoz0OssrGRk2Czp0CXSEp21nSQX3vbGOV1ftIiEmkt9fPpwrR6cQFho8tebBRAm3iIiIyCHOwfaPvJ7ZhwbTxPX2uoxkzIZu/QMd4Rkpq6jlb0s28fiH2wgJgW9PGsht5+uByOamv10RERGRkq1eXfbKeVC6HSI6eb2yM66FPucEVZeRo6mp8zHnk+389Z2NlFXWcuWoFL43JY2kOD0Q2RKUcIuIiEj7VFXmrWKvnAc7PgYMUsfDBT+FITMgIjrQEZ4x5xwLc3fz+wXr2FZcwbkDvAcih/bUA5EtSQm3iIiItB/1dbBlCaycC+teg7oqiB8Ek+6BEddAXHKgI2wy2TtL+c1ra/jvtn0MTOzEYzePYcKgBKyVj5Fvi5Rwi4iISNu3Z42XZK96Fg7s9h54HHmDN/2x5yhoQ0nozpIK7l+4nn+vLCC+UwS/vWw4V2fpgchAUsItIiIibdPBvV6v7JXzYNdKCAmDgVO9hx8HTYWwyEBH2KTKKmv5n8WbeMz/QOS3Jg7gtvH96aQHIgNOnwERERFpO+qqYcMbXr/sTW+Brw56ZML0+yH9CoiOD3SETa623secpd4DkaWVtVw+MoW7pg6iR1yHQIcmfkq4RUREJLg5B/nLvX7ZOc9DVSnE9ICzv+F1GUkcEugIm4VzjjfX7OH3C9axde9BvtC/Gz+5aAjpycE1Ur49UMItIiIiwaksz9/K72ko3ghhUTB4hleXnXoBhIQGOsJmsyqvlF+/tpb/bC1hQGInHv1SFhekJeqByFZKCbeIiIgEj+oDsPbfXl321vcAB72/AOd8G4bOgqi23e4ub18Ff1y4npeyC+gWHcGvZ6Uze0wvPRDZyinhFhERkdbN54Nt73sr2WtehtqD0KUvTPiR18qva79AR9js9lfV8j+LN/Poh1sx4BsX9Odr4/sTExUe6NDkJCjhFhERkdZp7yavld/K+bA/DyJjYfgVXl1277PbVCu/Y6mt9zHvPzv4y9sbKTlYw+Ujk7lraho9O+uByGCihFtERERaj8p9kPOCVzKS91+wEOg/ESb/AgZfDOHtI9F0zvH22kJ+t2AtW4oOMi61Kz+7eKgeiAxSSrhFREQksOprYdPbXpK9fgHU10DCEJj8Sxh+NcT2CHSELWp1Xhm/eX0NS7eUkJoQzf/dmMWkIXogMpgp4RYREZGW5xzsXuXVZa9+Fg4WQcdukHWLVzLSI6NdlIw0VlBayR8WrufFFfl0jY7gVzOHMXtsb8L1QGTQU8ItIiIiLad8D6x+xhtMU5gLIeGQNg0yroOBkyG0/T0EWF5Vy9+XbOaRD7bigNsn9Of2Cf2J1QORbYYSbhEREWletZWw/nUvyd78DjgfJI+Gi/7oTX/s2DXQEQZEXb2Pef/dyV/e2kDxwRpmZfbkrqlppHTpGOjQpIkp4RYREZGm5xzs/MSb/pj7ElSXQWwynPMdr2QkYVCgIwwY5xyL1hXy29fXsrnoIGP7deWxi4cwIqVzoEOTZqKEW0RERJrOvu3+6Y/zYN9WCO8IQy6FjNnQ7/w2Pf3xZOTkl/Hb19fy0eZiUuOjefiG0Uwe2l0PRLZxJ0y4zexRYAZQ6JxL92/7A3AJUANsBm52zpWa2WTg90CEf9/3nXOL/OeMBh4HOgCvA3c455yZRQJPAKOBYuAa59y2prxJERERaUZV+72BNCufhu0feNv6ngfnfx+GXgqRMYGNrxUoKK3kj296D0R27hDOLy4dxnVn6YHI9uJkVrgfBx7CS4oPeQv4sXOuzszuA34M/BDYC1zinCsws3RgIZDsP+fvwK3AUryEexqwAPgysM85N8DMZgP3Adec6Y2JiIhIM/LVw5bFXpK99lWoq4Su/eGCn0HGNdC5d6AjbBUOVNfxv0s288/3t+CAW89P5RsXDNADke3MCRNu59x7Ztb3iG1vNnq5FLjSv31Fo+25QJR/BbsrEOuc+xjAzJ4AZuEl3DOBe/3nPAc8ZGbmnHOncT8iIiLSnPas8cpFVj0DB3ZDVBxkXut1GUnJanet/I6lrt7H/GU7+fNbG9h7oIZLM3ry/alp9OqqByLbo6ao4b4FmH+U7VcAK5xz1WaWDOQ12pfH4ZXvZGAngH/FvAzohrda/hlmdiveKjm9e+t/ziIiIi3iQBHkPOcl2rtWQkgYDJjsJdqDpkFYZKAjbDWccyxZX8RvX1/LxsIDjOnbhf+7aQyZvToHOjQJoDNKuM3sp0AdMOeI7cPwSkOmHNp0lNPdSez77EbnHgYeBsjKytIKuIiISHOprYINb3hJ9sa3wNVDj0yYdp/Xyq9TQqAjbHVyC7wHIj/cVEzfbh353y+OZuowPRApZ5Bwm9lNeA9TTmpc/mFmKcCLwI3Ouc3+zXlASqPTU4CCRvt6AXlmFgbEASWnG5eIiIicJucg77/+Vn4vQFUZxPSAL3wTRsyG7kMDHWGrtLusij++uZ7nP80jrkM491wylOvP6kNEmB6IFM9pJdxmNg3vIcnxzrmKRts7A6/hPVD54aHtzrldZlZuZuOAT4AbgQf9u18BbgI+xqsFX6T6bRERkRa0bzusmu+tZpdsgbAOMOQSr5Vf6oR238rvWA5U1/Hwu5t5+P0t+Hzw1fO8ByLjOuiBSPmsk2kLOA+YAMSbWR5wD15XkkjgLf+vSZY6574GfBMYAPzczH7uv8QU51whcDuH2wIu8L8BPAI8aWab8Fa2ZzfJnYmIiMixHauV33nfg6Ez1crvCPU+x5aiA2TvLGVlXikrd5axdtd+6nyOSzJ68gM9ECnHYcG6mJyVleWWLVsW6DBERESCx7Fa+WVeCyPUyu8Q5xy7yqpYubOU7LxSVu4sZXVeGQdr6gHoFBnGiJQ4Mnp1ZuqwJD0Q2Q6Y2XLnXNbpnq9JkyIiIm3d51r5dYbM67wR62rlR1lFrX/V2lu9zt5Zxt4D1QCEhxpDe8RyxegUMlI6k9GrM6nx0YSEtO+/Mzk1SrhFRETaoqO18hs4xavLbset/Kpq68kt2N+QXK/KK2Pr3oMN+/snRHP+oHgye3UmI6Uzg3vEEBmmGnY5M0q4RURE2orjtfIbfiVExwc6whZV73NsKjzQkFyvzCtl3a5y6nxeOW1SbBQZveK4KstbvR6eEqcJkNIslHCLiIgEs0Ot/FbOg5znP9vKL+NaSBwS6AhbhHOOAn/d9cqdpWTvLGV1fhkV/rrrmKgwMlI6c+v5qWT4V6+T4qICHLW0F0q4RUREgtG+7V5N9sp5ULL5cCu/zGuh3/g238qvtKKGlXllDQn2yrxS9h6oASAiNIQhPWO5anSKl1z36ky/bqq7lsBRwi0iIhIsjtvK79I228rPq7suI3tnWUNyvb3YGwNiBv0TOjF+UCKZvbzOIYOTYjV0RloVJdwiIiKt2bFa+U38WZts5Vfvc2wsLPeXhZSxKq+UdbvLqffXXfeIiyIjpTOzx/Qmo1ccw5PjiFHdtbRySrhFRERao3bQys85R96+yoZuIdk7S8k5ou46s1dnbh/fv6HvdfdY1V1L8FHCLSIi0lq08VZ++w7WNExpPNT3uvigv+46LIRhPWO5OqsXGb3iyEjpTF/VXUsboYRbREQkkBpa+T0Nm94CX53Xym/6/ZB+RdC28qusOVR3XdrwcOOOksN11wMSOnHB4EQyenUmM6UzaUkxqruWNksJt4iISEs7Viu/s7/prWYHYSu/nSUVfLhpb8Okxg17Dtdd94yLIqNXZ647q3dDv+tOkUpBpP3Qv3YREZGWcmQrv/COXiu/jNlB2cpv74FqXlu1i5ez8/l0RykAsVFhZPTqzIVD+pOR0pkRveJIjFHdtbRvSrhFRESaU9V+WPsKZM9rE638DlTX8Wbubl7OLuCDTXup9zkGJ8Xww2mDmTKsO6nx0VgbeKBTpCkp4RYREWlqvnrYssRbyT7Uyq/bgKBt5VdT5+O9DUW8lJ3P22v3UFXrI7lzB247P5WZmcmkJQXXfxpEWpoSbhERkaaydyNkz/EegCzfdbiVX+Z1kDw6qFr5+XyO/24r4eWVBby+ehelFbV06RjOVaN7MTOzJ6N6d1EHEZGTpIRbRETkTFSVQc4LkD0X8v4DFgoDJ8P0+4Kyld/aXft5KTuff2cXUFBWRYfwUKYM686szGTOHRhPeKg6iYicKiXcIiIip+pQyUj2XFj3KtRVQcIQmPJrGH41xHQPdISnZGdJBa+sLOCV7ALW7yknLMQ4f1ACP5w+mMlDu9MxQumCyJnQV5CIiMjJ2rsJVs71Skb253slIyNvgJHXe72zg6hkpORgDa+t3sXLK/JZtn0fAFl9uvCrWelclJ5Et07BtTIv0pop4RYRETmeqjLIfdFbzd75CVgIDJgMU38LadODqmSkoqaOt9bs4eXsAt7bUESdzzGoeye+PzWNSzN60qtrx0CHKNImKeEWERE5ks8HW9/1kuy1//a6jMSnweRfel1GYpICHeFJq6338f7GIl7OLuDN3D1U1tbTMy6KL5/Xj1mZyQxOilEbP5FmpoRbRETkkOLNXiu/7HmwPw+i4rwOIyOvh56jgqZkxOdzfLpjHy9l5/Paql3sq6ilc8dwLhuVzMyMnozp21UdRkRakBJuERFp36rLD5eM7PjYKxnpPwmm/ArSLoLw4JmSuH53OS9n5/NydgH5pZVEhYcweWgSMzN6cv6gBCLC1GFEJBCUcIuISPvj88G29/0lI69AbQXED4IL7/VKRmJ7BjrCk5ZfWskr2QW8nJ3Put3lhIYY5w6I566pg5g8NIlOkfpRLxJo+ioUEZH2o2SLVy6ych6U7YTIOMiYDZnXB9VgmtIKf4eR7AL+s7UEgFG9O/OLS4dx8YgexKvDiEirooRbRETatupyWPOyt5q9/UPAoP9EbzV78MUQ3iHQEZ6Uypp63lq7h1ey83l3QxG19Y7+CdF8b/IgZmYm07ubOoyItFZKuEVEpO3x+bzkOnuul2zXHoRuA2DS3TBiNsQlBzrCk1JX7+ODTXt5ObuAhbm7qaipJyk2ipvP6celGT0Z1jNWHUZEgoASbhERaTv2bfOXjMyF0h0QGQsjrvJKRlLGBEXJiHOOT3eU8kp2Pq+u2kXxwRpio8K4NKMnMzOTGduvK6HqMCISVJRwi4hIcKs+4D34mD3XexASg9QJMPFur2QkIjhKLTYVlvPSigJeXpnPzpJKIsNCuHBId2Zm9mR8WgKRYaGBDlFETpMSbhERCT7OwfaPIHsO5L7klYx0TYWJP4OMayEuJdARnpRdZZX8e2UBL60oYM2u/YQYnDMgnjsmDWLqsO7ERIUHOkQRaQInTLjN7FFgBlDonEv3b/sDcAlQA2wGbnbOlZpZN+A5YAzwuHPum42uMxp4HOgAvA7c4ZxzZhYJPAGMBoqBa5xz25rsDkVEpO3Ytx1WPu2VjOzbBhExMPwKr2Sk11lBUTJSVlHLgpxdvJSdzydbS3AOMnp15p5LhnLxiB4kxgRP328ROTkns8L9OPAQXlJ8yFvAj51zdWZ2H/Bj4IdAFfBzIN3/1tjfgVuBpXgJ9zRgAfBlYJ9zboCZzQbuA6453RsSEZE2puagN149ew5sfQ8w6Hc+TPgJDLkkKEpGqmrreWdtIS9n57NkfRE19T5S46P5zqRBzMzsSd/46ECHKCLN6IQJt3PuPTPre8S2Nxu9XApc6d9+EPjAzAY0Pt7MegCxzrmP/a+fAGbhJdwzgXv9hz4HPGRm5pxzp3E/IiLSFjgHO5ZC9lNeyUjNAejSDy74GWRcA517BzrCE6qr9/HR5uKGDiMHqutIjInkhrP7MCszmfRkdRgRaS+aoob7FmD+CY5JBvIavc7zbzu0byeAf8W8DOgG7D3yImZ2K94qOb17t/5vtiIicopKd3olI9lzYN9WiOgEw2Z5JSO9z271JSN19T7+s7WEN3J38/rq3ew9UE1MZBgXDU9iZmYy41K7qcOISDt0Rgm3mf0UqAPmnOjQo2xzJ7Hvsxudexh4GCArK0sr4CIibUFNBax71Uuyt7wLOOh7Hoz/IQy9FCJad7lFVW09H2zcyxu5u3l77R5KK2qJCg/hgrREZmb2ZEJaIlHh6jAi0p6ddsJtZjfhPUw56STKP/KAxo+MpwAFjfb1AvLMLAyIA0pONy4REQkCzsHOT7wkO+dFqCmHzn1gwo+9Uetd+gQ6wuMqr6pl8foiFubsZvH6Qipq6omJCmPykO5MGZbE+EEJdIhQki0intNKuM1sGt5DkuOdcxUnOt45t8vMys1sHPAJcCPwoH/3K8BNwMd4teCLVL8tItJGleX5S0bmQslmCI/2l4xcB72/ACEhgY7wmEoO1vD2mj28kbubDzbupabeR3ynSGaNTGbasCTGpXYjIqz1xi8igXMybQHnAROAeDPLA+7B60oSCbzlf+BjqXPua/7jtwGxQISZzQKmOOfWALdzuC3gAv8bwCPAk2a2CW9le3bT3JqIiLQKtZWw7jVY8RRsWQI46HMunH8XDLkUIjsFOsJjKiit5M3c3byRu5v/bC3B5yClSwduPLsP09KTGNm7i2qyReSELFgXk7OystyyZcsCHYaIiByNc7DzP/7BNC9C9X6I6+2tZGfMhq79Ah3hMW0pOsAbubtZmLOblXllAAzq3olpw5KYmp7E0B7qLiLS3pjZcudc1umer0mTIiLSdEp3wqqnIXuev2SkIwydBZnXeqvarbBkxDlHbsH+hpXsDXsOAJCREscPpqUxdVgS/RNa7yq8iLR+SrhFROTM1FQcMZjG32XkvO95XUYiYwId4ef4fI5Pd+zjjRwvyc7bV0mIwdh+Xbn3kqFMGZZEz84dAh2miLQRSrhFROTUOQc7PvaXjLzcqMvIj/xdRvoGOsLPqanzsXRLMW/k7uatNXsoKq8mIjSEcwfG8+2JA5k0JJFunSIDHaaItEFKuEVE5OSV7jjcZaTxYJqM67zBNK2sZKSypp73Nnrt+95eu4f9VXV0jAjlgrREpqYncUFaAjFR4YEOU0TaOCXcIiJyfNUHDpeMbHsfMOh3vreaPeSSVjeYpqyylsXrCnkjZzdLNhRSVeujc8dwpgxLYtqwJM4dGK9BNCLSopRwi4jI5/l8sP1DWDkPcl+C2oPQNRUu+BlkXAOdewc6ws8oKq/mrTV7WJi7m48276W23tE9NpKrs3oxbVgSY/t1JSy0da2+i0j7oYRbREQOK9nqlYysnOuVj0TEwPArIPN66HUWtKJ2eHn7KliYu4eFObv57/YSnIM+3Tpyyzn9mJqeRGZKZ0LUI1tEWgEl3CIi7V11Oax52avL3v4hYJA6ASbeDYMvhoiOgY6wwabC8obOIjn5+wEYnBTDHZMGMi09ibTuMeqRLSKtjhJuEZH2yOfz6rGz58LaV6C2AroNgEl3w4hrIC4l0BECXo/s1fllvJGzm4W5u9lcdBCAUb0785OLBjN1WBJ9urWuGnIRkSMp4RYRaU+KN3t12SufhrKdEBnnJdiZ10NKVqsoGan3Of67rYSFubt5M3cP+aWVhIYYZ6d240tf6MuUYUl0j40KdJgiIidNCbeISFtXVeY9+Jg9F3YuBQuB/hNh8i8g7SIID/yAl+q6ej7aXMzCHK9HdvHBGiLDQjhvYAJ3Th7EhUMS6dwxItBhioicFiXcIiJtka8etr7rLxl5FeoqIT4NLrzXW9GO7RnoCDlYXce7G4p4I2c3i9cVUl5dR6fIMCYOTmRaehLjByUQHakfUyIS/PSdTESkLdm70UuyV82H/fkQFQeZ13klI8mjAl4yUlpRwztrC3kjdzfvbSiius5H1+gILh7Rg6nDkvjCgG5EhqlHtoi0LUq4RUSCXWUp5L4A2fMg7z9eyciAC2Hqb2DQdAgPbL1z4f4qFq7x2vct3VJMnc/RIy6Ka8f2Zlp6Ell9uqhHtoi0aUq4RUSCka8eNi/2+mWvfRXqqyFhCEz+FYy4GmKSAhpe3r4Kr31fzm6W79iHc5AaH82t56cyLT2J4clxat8nIu2GEm4RkWBStP5wyUj5LujQBUbf5JWN9MgMaMnI1r0HWZCzizdydrMqrwyAIT1iufPCQUxLT2JgYicl2SLSLinhFhFp7Sr3Qc7zXqKdvxwsFAZOgen3w6CpEBYZkLCcc6zfU86C1d5K9vo95QBk9OrMj6YPZnq6emSLiIASbhGR1qm+Dja/4yXZ61+H+hrong5TfwvDr4JOiQEJ69AgmgX+cpGtew9iBmP6duWeS4YydVgSPTsHvs2giEhrooRbRKQ12bPGq8te9Qwc2AMdu0HWl/0lIyMCEpLP51i+Yx8LVnvTHg8NovlC/2585bx+TBmaREJMYFbZRUSCgRJuEZFAqyiB1c9B9hzYlQ0hYTBompdkD5gMYS0/8KWu3scnW0tYkLOLhbl7KCqvJiI0hPMGxvOdCwcyeWh3DaIRETlJSrhFRAKhvhY2ve0l2evfAF8tJI2AaffB8CshOr7FQ6quq+ejTcUsyNnFW2v2sK+ilg7hoVwwOIFp6T24IC2BmKjwFo9LRCTYKeEWEWlJe3JhxRxY/QwcLILoBDjrNsi4FpLSWzycypp63t1QyIKc3Sxa6017jIkMY9KQRKal92D8oAQ6RGgQjYjImVDCLSLS3CpKvC4jK57yl4yEQ9p0f8nIhRDasqvG5VW1LFpXyBs5u1myvojK2nq6dAxn+vAkpqf30LRHEZEmpoRbRKQ5+Ophy2JvNXvdq16XkaTh/pKRqyC6W4uGU1pRw1tr9vBGzm7e37iXmnofCTGRXDk6henpSYzt11XTHkVEmokSbhGRplS82avLXvk07M/3BtNk3QKZ17d4l5HC8irezPWS7I+3FFPvcyR37sANZ/dhenoSo3p3ISREg2hERJqbEm4RkTNVfQDWvOStZu/4CCzEKxWZ+luvdKQFB9MUlFY2jFT/7/YSnIN+/pHq0zVSXUQkIJRwi4icDudgx8deXXbuS1B7ELoNgEn3eA9AxvZosVC27T3oH0Szi5X+keqDk2K4Y9JApqUnkdY9Rkm2iEgAKeEWETkVZfneYJrsuVCyBSJiYPgVkPlF6DUWWiCxdc6xsfAAC1bvZkHOLtbt9kaqj0iJ4wfT0pg2LInUhE7NHoeIiJycEybcZvYoMAModM6l+7f9AbgEqAE2Azc750r9+34MfBmoB77tnFvo3z4aeBzoALwO3OGcc2YWCTwBjAaKgWucc9ua7hZFRM5QbRWsf80rGdm8CHDQ9zw4/wcw9FKIiG72EJxz5BbsZ0HOLhbk7GZLkTdSfXTvLvzs4iFMS08ipUvHZo9DRERO3cmscD8OPISXFB/yFvBj51ydmd0H/Bj4oZkNBWYDw4CewNtmNsg5Vw/8HbgVWIqXcE8DFuAl5/uccwPMbDZwH3BNU9yciMhpc85r4bfiKW8KZFUpxPWC8T/wSka69mv2EHw+x4qd3kj1N3J3k7fPG6k+LrUrN5/Tj6lDu5MYG9XscYiIyJk5YcLtnHvPzPoese3NRi+XAlf6358JPO2cqwa2mtkmYKyZbQNinXMfA5jZE8AsvIR7JnCv//zngIfMzJxz7jTvSUTk9B3cC6vme6vZhbkQFgVDLvG6jPQbDyHN2zqvrt7Hf7aV8EbObhbm7mbP/mrCQ41zB8Tz7YkDuXBod7pGa6S6iEgwaYoa7luA+f73k/ES8EPy/Ntq/e8fuf3QOTsB/CvmZUA3YG8TxCYicmL1dbDpLW81e8Mb4KuD5NFw8Z8g/Qro0LlZP3xNnY+PNu/ljZzdvLlmDyUHa4gKD2H8oASmp/dg4pBEYjVSXUQkaJ1Rwm1mPwXqgDmHNh3lMHec7cc752gf71a8shR69+59SrGKiHxO4TrIfgpWzoeDhd6Y9XG3e6vZiUOa9UNX1dbz7oYi3sjZzdtr91BeVUd0RCgTh3RnenoSE9IS6Bih59pFRNqC0/5ubmY34T1MOalR+Uce0KvRYSlAgX97ylG2Nz4nz8zCgDig5Ggf0zn3MPAwQFZWlkpOROTUVZX5x6zPgfxlEBIGg6Z5SfbAyc06Zr2wvIr3N+xl0bpCFq8vpKKmnrgO4UwdlsT09CTOGRBPVLhGqouItDWnlXCb2TTgh8B451xFo12vAHPN7E94D00OBP7jnKs3s3IzGwd8AtwIPNjonJuAj/FqwRepfltEmpTPB9ve80pG1v4b6qogcag3mGb41dApoVk+bG29j0+37+PdDUW8u6GI3IL9AMR3imTWyGSmpycxLrUb4RqpLiLSpp1MW8B5wAQg3szygHvwupJEAm/5hyksdc59zTmXa2bPAGvwSk2+4e9QAnA7h9sCLvC/ATwCPOl/wLIEr8uJiMiZ27fN65edPQ/KdkBUHIz8orea3XNks/TMLiit9BLs9UV8uGkv5dV1hIYYo3t34ftT0xg/KIGhPWI1Ul1EpB2xYF1MzsrKcsuWLQt0GCLS2tRUwNpXvNXsbe8DBv0v8JLswTMgvGnb6FXX1fPfrft4d0Mh724oYsOeAwD0iItiQloC4wcl8IUB8XroUUQkiJnZcudc1umerydyRCT4OQd5/4UVT0LOi1BTDl36wcSfeT2z41JOfI1TsL34YMMq9kebi6msrSciNISx/bpy1ehejE9LYGBiJ41TFxERQAm3iASz/btg1dPeA5DFGyE8GobN8laz+3yhyUpGKmvqWbqlmHc3FLFkfSHbir1HV3p37chVWSmMH5TAuNRuREfqW6qIiHyefjqISHCpq4ENC7ySkU1vg/NB77Ph3O/A0JkQGXPGH8I5x+aiAyxZ7z3s+MnWEmrqfESFh3B2aje+9IW+TEhLpG988490FxGR4KeEW0SCw65VkD0HVj0DlSUQ0xPOvdNbze7W/4wvX15Vy0ebi1myvoj3NhSRX1oJwIDETtwwrg8T0hIY07er2vaJiMgpU8ItIq1XRQmsftZbzd69CkIjYPDFkPlF70HIkNNPfp1zrN1VzpINhby7vojl2/dR53NER4RyzoB4vnHBAM4fFE9Kl45NeEMiItIeKeEWkdbFVw+bF3kPQK5fAPU10CMDLvqjN2a9Y9fTvnRpRQ3vb9zb0Be7qLwagCE9Yvnq+amMH5TAqN5diAhTX2wREWk6SrhFpHXYu8k/Zv1pKN8FHbvBmK94JSNJ6ad1SZ/PsSq/jHfXF/HuhkKyd5bicxDXIZzzBsYzflAC5w9KoHts07YKFBERaUwJt4gETtV+WPOS12Vk51KwUG+8+vT7vXHrYRGnfMmi8mre3+itYL+/cS8lB2swgxEpnfnmxIGMH5RARkocYZruKCIiLUQJt4i0LF89bFnsrWSvfRXqKiE+DSb/EkZcAzFJp3S5unofK3aW+lexi1idXwZAfKcIJgxKYHxaAucOiKdbp8jmuBsREZETUsItIi2jcB2snOt1GSnfBVGdIfM67y159Cn1zN5VVsl7Gw6vYpdXeePTR/XuzF1TBjEhLVHj00VEpNVQwi0izedgMeQ8B9lzYVc2hITBgMkw/T5/ycjJrTpX19WzfNs+/+CZItbvKQcgKTaKi9J7MCHNG58e10Hj00VEpPVRwi0iTauuBjYu9EpGNiwEXy0kjYBpv4f0K6FTwkldZmdJBUs2FPHu+kI+2lxMRU094aHGmL5d+cnowYwflMig7hqfLiIirZ8SbhE5c85BwQpYOQ9WP+cNpunUHc66zSsZ6T7shJeoqvXGpx8aPLNl70EAUrp04PJRyUwYlMjZ/TU+XUREgo9+conI6dtfAKvme6vZResgNNI/mOY6SL0AQo/9LeZgdR1rd+0ne2cp723cyydbiqmu8xEZFsK41G7ccHYfxg9KoF98tFaxRUQkqCnhFpFTU1MB6171VrO3LAHng17jYMZfYNhl0KHz504pq6wlt6CM3Pz95BSUkZNfxpa9B3HO25+aEM31Z/VhfFoCZ/XT+HQREWlblHCLyIn5fLDjY6/LSO7LUFMOcb3hvLsgYzZ0699waPGBanIK9pOTX0ZuQRk5+fvZUVLRsL9nXBTDkuO4NCOZ9ORY0pPjNHhGRETaNCXcInJsJVu8cpGVT0PpdojoBENnQea1uN5ns6e8lpz8MnJWbCAnfz+5BWXsKqtqOL13146kJ8cye2wv0nvGMaxnrPphi4hIu6OEW0Q+q6oMcl/ySkZ2fAwYLnU8JWPvYnmHc1hZWEvOov3kFixi74EawGuhnRofzdh+Xb3EOjmWYT3iiOuoNn0iIiJKuEUE6uu8euyVc3HrXsPqqijvlMqylK/zfN0XeG9LBPvX1AHrCA0xBiZ2YkJaIuk9vZKQIT1i1T1ERETkGPQTUqQdq9uVw/6lT9Bh3Qt0qC6i3GL4t28882vOYWVVfyL2hTK4RwwXj4jz6q17xpGWFKOHGkVERE6BEm6RdqK6rp6New6wcctWwte9wJA9r9G/bhMxLpQlvkz+zQ3s6TGewcnxXJ8cx+96xjGweyfCQ0MCHbqIiEhQU8It0gZV1tSzdvd+cvO9LiHr8veSUvQes+w9ZoRkE271bAkfwMJed1I/7AoGpfblz/GdCA1Rv2sREZGmpoRbJMiVV9WypmA/OQX+BLugjE2FB/A5R4Zt5trID/lZyEfEhJVTFZlAxdCvEnPWjaQmDSM10MGLiIi0A0q4RYLIvoM15BYcHh6TW7Cfrf4R6ACJMZGc372Gn3f+kJElC+hUvgUXFoUNvhgyriMqdQJRx5n+KCIiIk1PP3lFWqnC8ipvMqN/1Tonfz/5pZUN+5M7dyA9OZbLRyYzons4mQc+IG7Ds7DlXcBB77NhwnewYbMgKi5g9yEiItLeKeEWaQV8Pseq/DLeXV/EyrxScvLLKCyvbtjfLz6akb07c8PZfRoGyHTpEAbbP4SVj8EnL0HNAejcG8b/wJv+2FUFIyIiIq2BEm6RANlfVcv7G/ayaF0h724oZO+BGsxgYGInzh0Qz7DkONJ7xjK0ZywxUY0GyBRvhk/+AauehtIdEBEDw2ZBxnXeqnaIuoqIiIi0Jkq4RVqIc47NRQdYtK6QResKWbZtH3U+R1yHcMYPSmDi4ETGD0qgS3TE50+uLIXcF73pjzs/AQxSJ8DEn8PgGRDRsYXvRkRERE6WEm6RZlRVW88nW0tY7E+yd5RUADA4KYavnp/KxMGJjOzVmbCj9bqur4PNi7wke91rUF8N8Wlw4b0w4hqI7dmyNyMiIiKn5YQJt5k9CswACp1z6f5tVwH3AkOAsc65Zf7tEcA/gCzAB9zhnFvi3zcaeBzoALzu3+fMLBJ4AhgNFAPXOOe2NdkdirSwXWWVLF5XxKJ1hXy4aS+VtfVEhYfwhf7x3Hp+KhcMTiS5c4djX2BPLmTPhdXPwoE90KELjL7Jq8vuOQpMvbJFRESCycmscD8OPISXFB+SA1yOl1w39lUA59xwM0sEFpjZGOecD/g7cCuwFC/hngYsAL4M7HPODTCz2cB9wDWnfUciLaze58jeuc9fKlLE2l37Aa+LyJWjU5g4OJGz+3c7/jj00h2Q8wLkPAe7V0NIGAycCpnXen+GHaXMRERERILCCRNu59x7Ztb3iG1rAezzK21DgXf8xxSaWSmQZWY7gVjn3Mf+854AZuEl3DPxVssBngMeMjNzzrnTuiORFlBWUcu7G4tYvK6QJesL2VdRS2iIMbpPF340fTATBycyMLHT0b5GDivfDbkvQc7zkPcfb1vPUTD9fki/EqK7tci9iIiISPNq6hrulcBMM3sa6IVXJtILr7wkr9FxeUCy//1kYCeAc67OzMqAbsDeIy9uZrfirZLTu3fvJg5d5Nicc2zY4z3wuHhdIct37KPe5+jSMZwJaYlMHJzI+QMTiOsYfvwLVZTA2le8JHvbB+B80D0dJt0Nwy6Hrv1a5oZERESkxTR1wv0oXl33MmA78BFQBxxtme/QCvbx9n12o3MPAw8DZGVlaQVcmlVVbT0fbd7rT7KLGobODO0Ry+3j+3PB4EQye3UmNOQENdVV+2H9616SvXkR+Oqga384//tekp04uAXuRkRERAKlSRNu51wdcOeh12b2EbAR2AekNDo0BSjwv5+HtwqeZ2ZhQBxQ0pRxiZys/NLKhlXsjzbvparWR8eIUM4ZEM83Jw7ggrREkuKiTnyhmgrYuNBLsje86XUYiesFZ38D0q+ApBF6+FFERKSdaNKE28w6AuacO2hmk4E659wa/75yMxsHfALcCDzoP+0V4CbgY+BKYJHqt6Wl1NX7+HRHaUOSvX5POQC9u3Zk9pjeTBycyFmpXYkMO84Djw0Xq/ZWsHOeh3WvQ+1B6NQdsm72kuyUMUqyRURE2qGTaQs4D5gAxJtZHnAP3gr0g0AC8JqZZTvnpgKJwEIz8wH5wA2NLnU7h9sCLvC/ATwCPGlmm/zXnX3mtyVybPsO1vDuhiL/hMciyiprCQsxxvTtyk8vGsIFgxPpnxB9/AceD6mvg23veUn22n9DVZnXxm/EVV6S3eccCDmJZF1ERETaLAvWxeSsrCy3bNmyQIchQcA5x9pd5Sxe7w2fWbFjHz4H8Z0iGh54PHdgPLFRJ3jg8RCfD3Yu9ZLs3JegYq83Xn3IDC/JTp0AoSd5LREREWn1zGy5cy7rdM/XpElpkypq6vhwUzGL/G37dpVVATA8OY5vThzIpMGJDE+OI+REDzwe4hwUrPCS7JwXoLwAwjrAoKkw/EoYMBnCT6K2W0RERNodJdzSZuwsqfAPnynk4y3F1NT5iI4I5byBCdx5YSIT0hJIjD3FpHjPGn+S/Tzs2woh4TDgQpj8S0ibBpExzXMzIiIi0mYo4ZagVVvvY9m2fQ2lIpsKDwCQGh/NDeP6MHFwImP6diUiLOTULly82T/18XkoWgsWAv3Gw3nf88pGOnRphrsRERGRtkoJtwSVvQeqWbLem/D43sYiyqvqCA81zurXjevGel1F+sZHn/qFS3dC7otekr0r29vW+wtw0R9h6CzolNCUtyEiIiLtiBJuadWcc+QW7G8oFVmZV4pzkBgTyUXpPbjA/8Bjp8jT+Kd8oPDwaPWdS71tPUfBlN/AsFkQl3K8s0VEREROihJuCbiq2noK91ezp7yKPfur2LO/msL9VRSUVfHJlmIKy6sxg4yUztx54SAmDk5kaI/Yk3/gsbGKEq99X87zsO19b7R64jCY+HNIvxy6pjb9DYqIiEi7poRbmk1tvY+i8mp276+i0J9INyTUjZLrssraz50bERZC99hIxvTtysTBiYxPSyC+U+TpBVJd7g2iaRitXusl1ufd5SXZiUPO8E5FREREjk0Jt5yyep+j+ED14QS6/PCqdOOEeu+Bms+dGxZiJMZEkhgbRb/4aMaldqN7bBSJMZF0j43yv0US1yH85AbPHEttJWx8E1Y/5/1ZVwWxKTDua5B+JfTI0NRHERERaRFKuKWBz+fYV1HjJdLlx16VLiqvxnfEvCQziO8USffYSHrERZHZuzPdY7zkuXtsFIn+P7t2jDi9UpCTUVcDWxb7R6u/BjUHIDoBRt3oH60+FkJOsWOJiIiIyBlSwt0OOOfYX1XnT5wPJ9ENCXV5FYX+hLq2/vOTR7tGRzSsQA9OivEn0FF0b7QqHd8pgrDQACSzvnqvFjvneVjzClSVQlRnr1Qk/Qrocy6E6p+5iIiIBI4ykSB3sLruqHXRXkJ9+EHEqlrf586NjQprSJjPSo323o+JPJxQx0aSEBNJZFhoAO7sOHw+yPvP4dHqBwshohMMvtg/Wv0CCIsIdJQiIiIigBLuoLF4fSFLNxcfTqj9q9IHqus+d2yH8FCS4ry66IyUzo3KOj67Kt0hopUl0sfjnNcfO+d5yHkR9udBWJQ3Wj39Chg4BcI7BDpKERERkc9Rwt3KVdXW88tX1zD3kx0NnTu6x0QxJCmW8YMiGx4y7B5zeFW6U2TYmT1w2JoUrj08Wr1ki3+0+iS48B5Im67R6iIiItLqKeFuxTYVlvPNuStYt7uc28ancteUNMIDUSfd0sryYfWzsOoZKMz1j1Y/H869EwbPgI5dAx2hiIiIyElTwt0KOed4bnked7+cS8eIUB6/eQwT0hIDHVbzqirzBtKsfBq2fQA4r6vI9Pu90eox3QMdoYiIiMhpUcLdyhyoruPnL+Xw4op8xqV25a+zR9I9NirQYTWPuhrY/A6smg/rF3i9srumwoQfw4irNPVRRERE2gQl3K1IbkEZ35q7gm3FB7nzwkF8c+IAQpurZ3WgOAd5y2DV05DzAlSWQMduXq/sEddA8mgNpBEREZE2RQl3K+Cc48ml2/n1a2vp0jGcuV8dx7jUboEOq2kVb/ZqslfNh31bvQ4jgy/2kuz+EyE0PNARioiIiDQLJdwBVlZRyw+eX8nC3D1ckJbAH6/KoFunyECH1TQO7vVWsVfNh/xlgHkPP57/fRhyCUTFBjpCERERkWanhDuAPt2xj2/NXcGe/VX89KIhfPncfs039ryl1FbC+te91exNb4OvDroPh8m/guFXQmzPQEcoIiIi0qKUcAeAz+d4+P0t/HHhepLionj2a2czsneXQId1+nz1XmeRVfO98eo15RDTE87+hlcy0n1YoCMUERERCRgl3C1s74FqvvfMSt7dUMRFw5P43eUjiOsQpPXLu3O8JHv1c1BeABExMHQmZFwDfc6BkCCaZCkiIiLSTJRwt6CPNu/lO09nU1pZy69npXP9Wb2DbyLkkUNpQsJgwGSY+htv8qPGq4uIiIh8hhLuFlDvc/z1nY08uGgj/eKjefzmsQztGUQPDFbth7WveKvZW9/HG0ozBi76Iwy7HKLbWEcVERERkSakhLuZ7S6r4o6nV/DJ1hKuGJXCL2cOIzoyCP7a62u9hx4/N5TmRzD8KujWP9ARioiIiASFIMj8gtfidYV879mVVNXW8/+uyuCK0SmBDun4GobSzIec572hNB26wsgbIGO2htKIiIiInAYl3M2gps7HHxau45/vb2VwUgx/u34U/RM6BTqsYzvaUJq0i7wOIwMmaSiNiIiIyBlQwt3EdpZU8M15K1i5s5QbxvXhpxcPISq8FXbrOFgMuf6hNHn/xRtKc56G0oiIiIg0MSXcTej11bv44fOrAPj79aOYPrxHgCM6Qm2lV4+96hnY9JZ/KE06TP4lpF8JccmBjlBERESkzTlhwm1mjwIzgELnXLp/21XAvcAQYKxzbpl/ezjwf8Ao/7WfcM79zr9vNPA40AF4HbjDOefMLBJ4AhgNFAPXOOe2Nd0tNr+q2np+9eoa5nyyg8xenXnw2pH06tox0GF5GobSPANrXv7sUJrhV0NSeqAjFBEREWnTTmaF+3HgIbyk+JAc4HLgH0ccexUQ6ZwbbmYdgTVmNs+fQP8duBVYipdwTwMWAF8G9jnnBpjZbOA+4JrTvqMWtqnwAN+c+ynrdpdz2/mp3DU1jfDQkECHBXtyvXKRVc9+dijNiKuh77kaSiMiIiLSQk6YcDvn3jOzvkdsWwscbWiLA6LNLAxvJbsG2G9mPYBY59zH/vOeAGbhJdwz8VbLAZ4DHjIzc86507ullvPc8jx+/lIOHSJCeezmMVyQlhjYgMryIec5bzV7T45/KM2FGkojIiIiEkBNXcP9HF4CvQvoCNzpnCsxsywgr9FxecChguFkYCeAc67OzMqAbsDeIy9uZrfirZLTu3fvJg795B2sruPnL+fwwqf5nNWvK3+dPZKkuKjABFO1H9b+G1Y9fZShNJdBdHxg4hIRERERoOkT7rFAPdAT6AK8b2ZvA0dr3nxoBft4+z670bmHgYcBsrKyArICvqZgP9+c9ylb9x7kjkkD+fakgYSGtHBv6vpa2PSOfyjN695Qmi79NJRGREREpBVq6oT7OuAN51wtUGhmHwJZwPtA46kvKUCB//08oBeQ5y9FiQNKmjiuM+ac46ml2/nVa2vp3CGcuV8Zx9n9W3CkuXOQv/zwUJqK4sNDaUZcAylZGkojIiIi0go1dcK9A5hoZk/hlZSMA/7inNtlZuVmNg74BLgReNB/zivATcDHwJXAotZWv11WWcuPnl/FgpzdTEhL4P9dlUG3TpEt88FLtngPPq6aDyWb/UNppntJdv9JEBbRMnGIiIiIyGk5mbaA84AJQLyZ5QH34K1APwgkAK+ZWbZzbirwN+AxvC4mBjzmnFvlv9TtHG4LuMD/BvAI8KSZbfJfd3aT3FkTWbFjH9+at4LdZVX8ePpgvnpeKiHNXUJSUQK5L3pJ9s5PAPM6i5x7Jwy9FKLimvfji4iIiEiTsVa2mHzSsrKy3LJly5rt+j6f45/vb+EPC9fTPTaKB68byajeXZrt41FXDRsWekn2hoXgq4WEwd5K9oirIS7lxNcQERERkSZnZsudc1mne74mTR5F8YFqvvfsSpasL2LasCTuu2IEcR3Dm/4DOQc7lnpJdu6LUFUK0Ykw9lbIuAaSRqguW0RERCTIKeE+wsebi/nO/BXsq6jlVzOH8cVxfY7Wb/zMFG+GlU97iXbpdgjvCINneKvZqRMgVJ8WERERkbZCmZ1fvc/x4KKNPPDORvp2i+bRL41hWM8mrJU+WOx1F1k1H/KXAQap42HCj2HIDIiMabqPJSIiIiKthhJuYM/+Ku54egVLt5Rw+chkfjUrnejIJvirqa2CDQtg5XzY9Bb46qB7Okz+FQy/EmJ7nvnHEBEREZFWrd0n3IvXF/K9Z1ZSWVPPH6/K4MrRZ/hwos8HOz7ySkbWvAzV+yGmB4y7HUbMhqT0pglcRERERIJCu024a+t9/HHhev7x3hYGJ8Xw0HWjGJDY6fQvWLTBG6++6lko2wHh0V4LvxHXQL/zISS06YIXERERkaDRLhPunSUVfGveCrJ3lnL9Wb35+YyhRIWfRkJ8oAhynvNWs3dlg4VA/4kw6ecw+GKIiG7y2EVEREQkuLS7hHvB6l384PlV4OBv143i4hE9Tu0CNRWw/nXv4cdN74Cr99r3Tf0tpF8JMd2bJ3ARERERCUrtJuGuqq3n16+t4amlO8hIiePBa0fRu1vHkzvZVw/bPvCS7DWvQE05xKbAOd/2SkYShzRv8CIiIiIStNpFwr256ADfnLuCtbv289Xz+vH9qYOJCAs58Yl71hyuyy4vgIgYGDrTG0rT51wIOYlriIiIiEi71uYT7hc+zeNnL+UQGRbCo1/KYuLgE5R8lO+G1c95ifbu1WChMOBCmPprSLsIwju0TOAiIiIi0ia02YT7YHUdd7+cy/Of5jG2X1f+OjuTHnHHSJZrDsLaV70ke8sScD7oOQqm3QfpV0CnhBaNXURERETajjaZcK/dtZ9vzP2UrXsP8u1JA/n2xAGEhR5R/uGr95LrVfO9ZLv2IMT1hnO/69VlJwwKSOwiIiIi0ra0qYTbOcecT3bwy1fX0LlDOHO+chZf6B//2YN2r/ba+K1+Dg7shsg4b+rjiGug99mqyxYRERGRJtVmEu6yylp+/MIqXl+9m/MHJfCnqzOI7xTp35kPq5+FVc9AYS6EhMHAKV6SPWgahEcFNngRERERabPaRMKdvbOUb879lF1lVfxo+mBuPS+VkNoDsOI5r2Rk63uAg5QxcNEfYdjlEN0t0GGLiIiISDsQ1Am3z+d45IOt3PfGOrrHRvHMV8cwui4bXvgtrHsN6iqhS18Y/wNvNbtb/0CHLCIiIiLtTNAm3HU+x5f/9V8Wry/kqwP2c1f3bCKfuxUOFkFUZ8i8FkbMhl5jwSzQ4YqIiIhIOxW0Cfe2Pfu4Ycsj/Lnbf+ictwV2RcCgqd5K9sApEBYZ6BBFRERERII34R7gtvO90Keh6zi44FswdBZ07BrosEREREREPiNoE24X0wO+/T507RfoUEREREREjilom05bTJKSbRERERFp9YI24RYRERERCQZKuEVEREREmpESbhERERGRZqSEW0RERESkGSnhFhERERFpRkq4RURERESakRJuEREREZFmdMKE28weNbNCM8tptO0qM8s1M5+ZZTXafr2ZZTd685lZpn/faDNbbWabzOwBMzP/9kgzm+/f/omZ9W362xQRERERCYyTWeF+HJh2xLYc4HLgvcYbnXNznHOZzrlM4AZgm3Mu27/778CtwED/26FrfhnY55wbAPwZuO+U70JEREREpJU6YcLtnHsPKDli21rn3PoTnHotMA/AzHoAsc65j51zDngCmOU/bibwL//7zwGTDq1+i4iIiIgEu7BmvPY1eMk0QDKQ12hfnn/boX07AZxzdWZWBnQD9h55QTO7FW+VHKC6cZlLOxEHlAU6iBame24f2ts9t7f7Bd1ze6F7bh/a4z0PPJOTmyXhNrOzgArn3KGE+Ggr1u4k9n12o3MPAw/7P8Yy51zW0Y5rq8zsYefcrSc+su3QPbcP7e2e29v9gu65vdA9tw/t9Z7P5Pzm6lIyG385iV8ekNLodQpQ0GhfLwAzC8P7X9NnSlikwb8DHUAA6J7bh/Z2z+3tfkH33F7ontsH3fMpMq+k+gQHeZ1DXnXOpR+xfQlwl3NuWaNtIcAO4Hzn3JZG2/8LfAv4BHgdeNA597qZfQMY7pz7mpnNBi53zl19EjG1uxVuEREREQk+JywpMbN5wAQg3szygHvwVqAfBBKA18ws2zk31X/K+UBe42Tb73a8jicdgAX+N4BHgCfNbJP/urNPMvYzWtoXEREREWkJJ7XCLSIiIiIip0eTJlsRM6s/YnBQ3+Mcu6Tx0KFgZGbOzJ5s9DrMzIrM7NVAxtUSzOwy//0PDnQszak9f44BzOxAoGMIlBPdexv5HtYuvo4bM7Of+gffrfL/nDor0DG1BDNLMbOXzWyjmW02s7+aWcRxjv+OmXVsyRibiv/f9P9r9PouM7s3gCE1u0b5V66ZrTSz7/pLpJuMEu7WpfLQ4CD/27ZAB9TMDgLpZtbB/3oykH8qF/A/aBuMrgU+4ORLqAAws9DmCafZnPHnWKQVO62v42BlZmcDM4BRzrkRwIX42/q2Zf7ZIC8ALznnBgKDgE7Ab45z2neAoEy4gWrgcjOLD3QgLehQ/jUM7+fURXgl1E2m1Sfc7XmFCMDMRpvZu2a23MwW+ocIHfJFM/vIzHLMbGzAgjwzC4CL/e83DEsCMLOx/vtb4f8zzb/9S2b2rJn9G3iz5UM+M2bWCTgHb8rqbP+2CWb2npm9aGZrzOx/D/3v2swOmNkvzewT4OzARX7aTudz/L6ZZTY67kMzG9GSQTcV/+f21UavHzKzL/nf32ZmvzCzT81sdVtbKT3evQe743wdH+tzfZGZrTOzD8zsgSD9LU8PYK9zrhrAObfXOVdwrJ9T/t9i/KUN/JyaCFQ55x4DcM7VA3cCt5hZtJn90f/1u8rMvmVm3wZ6AovNbHEA4z5ddXjPyd155A4z62Nm7/jv9R0z621mcf7vZYd+ZnU0s51mFt7SgTcF51wh3syXb5on1Mz+YGb/9d/3bYeONbMf+D/3K83s98e7bqtPuNuZDna4nORF/z/WB4ErnXOjgUf57P+oo51zXwC+7t8XjJ4GZptZFDACr4vNIevwut2MBO4Gftto39nATc65iS0WadOZBbzhnNsAlJjZKP/2scD3gOFAf+By//ZoIMc5d5Zz7oOWDrYJnM7n+P+ALwGY2SAg0jm3qsUibll7nXOjgL8DdwU6GDlpszj61/Hn+P/t/wOY7pw7F6/hQDB6E+hlZhvM7H/MbHw7+Tk1DFjeeINzbj9eR7avAP2Akf5V/znOuQfwWh9f4Jy7oKWDbSJ/A643s7gjtj8EPHHoXoEHnHNlwEpgvP+YS4CFzrnaFou2ifkbf4QAiXj/qS5zzo0BxgBfNbN+ZjYd7/vAWc65DOD+410zKBJuM+vk/5/UoVWgmf7tfc1srZn907y6mzft8K+ug1HjkpLLgDQgHXjLzLKBn/HZfubzAJxz7wGxZta5heM9Y/4kqi/eyufrR+yOA541b6Lon/G+6R3ylnMuWPu1X4uXhOL/81r/+/9xzm3xr57MA871b68Hnm/ZEJvOaX6OnwVm+H+Y34LX4aitesH/53K8vycJDsf6Oj6awcAW59xW/+t5xzm21XLOHQBG463+FQHzgdto4z+n8Ab0Ha3DhOF1Zvtf51wdQBD/XPoM/38ongC+fcSus4G5/vef5PDPqfl4E8bB+43P/OaOsQUcGsw4BbjR/+/7E7xp6APxSqoec85VwIk/98FS/1oFXOac229eTdFSM3vFv28gcK1z7qtm9gxwBfBUoAJtYgbkOueOVUZw5DeAYG058wrwR7z2k90abf8VsNg5d5l5D5AuabTvYEsF15TMrBveryfTzcwBoXift9c59uezyp+EB7NT+hw75yrM7C1gJnA1EMwP19Xx2cWNqCP2V/v/rCd4viefrBPde1A6ztfxKxz9fo82UTko+b8XLQGWmNlq4Bu0/Z9TuXi5RQMzi8Ub2reF4Lynk/EX4FPgseMcc+jeXwF+Z2Zd8f5Ttqh5Q2teZpaK9z25EO/r91vOuYVHHDONU/jcB8UKN97N/tbMVgFvA8lAd/++rc65bP/7bW2FaD2QYN6DKphZuJk1XuW9xr/9XLxfd5QFIMam8CjwS+fc6iO2x3H4AbsvtWhEzedKvF/H9XHO9XXO9QK24q0SjPX/mioE73MbjOUjx3I6n+P/Ax4A/hvkq0bbgaFmFun/9eykQAfUgtrqvR/r6xiOfr/rgFQ73HnqGoKQmaWZ2cBGmzKBtbT9n1PvAB3N7EZoeHj9/+H95u1N4Gvmf4Dfn3AClAMxLR9q0/F/330Gr6TikI84/JDw9fh/Tvl/+/Ef4K94gxKDdpHIzBKA/wUecl7v7IXA7Ydq0s1skJlF433ubzF/N5pGn/ujCpbVlOvxat5GO+dqzWwbh1cOqhsdV483WKdNcM7VmNmVwAP+b95heP/jzPUfss/MPgJi8X7tHpScc3l4X6RHuh/4l5l9lyD/33Ij1wJHPljxPN5gqI/9+4YD7wEvtmxozed0PsfOueVmtp/jr660Wv4fwNXOuZ3+376tAjYCKwIbWfNrB/d+rK/j6/ASlM/cr3Ou0sy+DrxhZnvxEpNg1Al40F8WUgdswisveZg2/HPKOefM7DLgf8zs53iLla8DP8HLOwYBq8ysFvgnXp3zw8ACM9sVxHXc4P3H4puNXn8beNTMvo9XVnRzo33z8coBJ7RYdE2ng79kJBzv3/aTwJ/8+/4PbzH3UzMzvPue5Zx7w7yH+5eZWQ2H/00cVasffGNel5KfAgOcc98yswvwfjD38x/SMHLezO4COjnn7g1IsCKnycwmAHc552YEOJRWw8x64v3qerBzzhfgcE6ZmWUA/3TOBWtnhtPWnu/9WMysk3PugP8H9t+Ajc65Pwc6ruZkZkvwvq8tC3QsIoHWqktKDq2S4D0Jm2Vmy/BWu9cFNDARaVb+X91+Avw0SJPtr+E9LPazQMfS0trzvZ/AV/0raLl4pVT/CGw4ItKSWvUKt1ZJRERERCTYtdoVbq2SiIiIiEhb0KpXuEVEREREgl2rXeEWEREREWkLWk3CbWa9zGyxf3Jkrpnd4d/e1czeMrON/j+7+Ld38x9/wMweOuJaS8xsvR0ek54YiHsSEREREWk1CTde38PvOeeGAOOAb5jZUOBHwDvOuYF4zed/5D++Cvg5cNcxrnd9ozHphc0cu4iIiIjIUbWahNs5t8s596n//XK86VXJeKOd/+U/7F/ALP8xB51zH+Al3iIiIiIirVKrSbgb84+/HYnXh7e7c24XeEk5cLLlIY/5y0l+7h80ICIiIiLS4lpdwm1mnfBG5H7HObf/NC9zvXNuOHCe/+2GpopPRERERORUtKqE28zC8ZLtOc65F/yb95hZD//+HsAJ67Gdc/n+P8uBuYAG54iIiIhIQLSahNtf9vEIsNY596dGu14BbvK/fxPw8gmuE2Zm8f73w4EZQE7TRywiIiIicmKtZvCNmZ0LvA+sBnz+zT/Bq+N+BugN7ACucs6V+M/ZBsQCEUApMAXYDrwHhAOhwNvAd51z9S10KyIiIiIiDVpNwi0iIiIi0ha1mpISEREREZG2SAm3iIiIiEgzUsItIiIiItKMlHCLiIiIiDQjJdwiIiIiIs1ICbeISJAzs3ozyzazXDNbaWbfNbPjfn83s75mdl1LxSgi0p4p4RYRCX6VzrlM59wwYDJwEXDPCc7pCyjhFhFpAerDLSIS5MzsgHOuU6PXqcB/gXigD/AkEO3f/U3n3EdmthQYAmwF/gU8APwemABEAn9zzv2jxW5CRKQNU8ItIhLkjky4/dv2AYOBcsDnnKsys4HAPOdclplNAO5yzs3wH38rkOic+7WZRQIf4k323dqS9yIi0haFBToAERFpFub/Mxx4yMwygXpg0DGOnwKMMLMr/a/jgIF4K+AiInIGlHCLiLQx/pKSeqAQr5Z7D5CB99xO1bFOA77lnFvYIkGKiLQjemhSRKQNMbME4H+Bh5xXMxgH7HLO+YAbgFD/oeVATKNTFwK3m1m4/zqDzCwaERE5Y1rhFhEJfh3MLBuvfKQO7yHJP/n3/Q/wvJldBSwGDvq3rwLqzGwl8DjwV7zOJZ+amQFFwKyWCV9EpG3TQ5MiIiIiIs1IJSUiIiIiIs1ICbeIiIiISDNSwi0iIiIi0oyUcIuIiIiINCMl3CIiIiIizUgJt4iIiIhIM1LCLSIiIiLSjP4/R0CxxSavSc0AAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 864x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "df['Money'][-n_obs:].plot(figsize=(12,5),legend=True).autoscale(axis='x',tight=True)\n",
    "df_forecast['MoneyForecast'].plot(legend=True);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "id": "27792a9d-d522-4b1e-a450-cda783a3207e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAtwAAAFICAYAAABjtimhAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/YYfK9AAAACXBIWXMAAAsTAAALEwEAmpwYAABabUlEQVR4nO3dd3yW5dn/8c+ZHUIGECCBkAQIJGwIkSVTWcpQXKi11drW0aHU6qO2dbb2aZ/2af1Znw5rrdqhooKyxYGKiCPskYRNEpKQEMgm4859/v647kBEZSa5cyff9+vFK8l1r+MiJPly5riO01hrERERERGR5uHn7QJERERERNoyBW4RERERkWakwC0iIiIi0owUuEVEREREmpECt4iIiIhIMwrwdgHnKzo62iYmJnq7DBERERFp4zZs2HDEWtv1fB/vs4E7MTGR9PR0b5chIiIiIm2cMebghTxeLSUiIiIiIs1IgVtEREREpBkpcIuIiIiINCOf7eH+KnV1deTm5lJdXe3tUqQFhISEEBcXR2BgoLdLEREREflabSpw5+bmEh4eTmJiIsYYb5cjzchaS3FxMbm5ufTu3dvb5YiIiIh8rTbVUlJdXU2XLl0UttsBYwxdunTRbzNERESk1WtTgRtQ2G5H9LkWERERX9DmAreIiIiISGuiwN0MnnjiCQYNGsTQoUMZPnw4n376abO91qOPPsrvfvc7AB5++GHeeeedZnstERERETl3beqiydZg/fr1LFu2jI0bNxIcHMyRI0eora1tkdd+/PHHW+R1RERERNqNugu/Xkwr3E0sPz+f6OhogoODAYiOjqZHjx4kJiZy//33M2rUKEaNGsWePXsAKCoq4uqrr+aiiy7ioosuYt26dYCzcn3rrbcyefJk+vTpw1NPPXXiNZ544gmSk5OZOnUqWVlZJ47fcsstvPbaawAkJibyyCOPkJqaypAhQ8jMzDzxetOmTSM1NZXbb7+dhIQEjhw50iJ/NyIiIiI+oaYctr8Or94Cv+17wU/XZle4H1u6g515ZU36nAN7RPDInEGnvc/06dN5/PHH6d+/P1OnTmX+/PlMmjQJgIiICD777DNefPFFFixYwLJly7j77rv58Y9/zPjx48nOzmbGjBlkZGQAkJmZyZo1aygvLyc5OZk777yTrVu38vLLL7Np0yZcLhepqamMHDnyK2uJjo5m48aN/OlPf+J3v/sdzz77LI899hiXXHIJDz74IKtWreKZZ55p0r8jEREREZ9UWQxZKyBjKexbA/W12LBulCVdAfz5gp66zQZub+nYsSMbNmxg7dq1rFmzhvnz5/PrX/8agBtuuOHE2x//+McAvPPOO+zcufPE48vKyigvLwdg1qxZBAcHExwcTLdu3Th8+DBr165l3rx5dOjQAYC5c+d+bS1XXXUVACNHjmTRokUAfPTRRyxevBiAmTNn0qlTp6Y8fRERERHfUZoLmcshYyn24DqMdVMe0oONUVeyoi6NJUfjOL4RFLi/xplWopuTv78/kydPZvLkyQwZMoQXXngB+OIYu4b33W4369evJzQ09EvP09CW0vCcLpfrS89zOg2Pb/xYa+15nJGIiIhI21BXuIuSDa8TkLWcTiXbANhnerGsbi5v1Y9iR3UC0a5gBsRGcFNyOCkxEVzzmwt7zTYbuL0lKysLPz8/+vXrB8DmzZtJSEhg27ZtvPLKKzzwwAO88sorjB07FnBaUJ5++mnuu+++E/cfPnz41z7/xIkTueWWW3jggQdwuVwsXbqU22+//azrGz9+PAsXLuT+++9n9erVHDt27PxPVkRERKQVK66oISOvjKI9n9Nx30r6H32fBHc2XYHN7j48a68nq9NkInoOICU2nAdiI0iJiaBrePCZnvqcKHA3sYqKCn70ox9RUlJCQEAASUlJPPPMMyxbtoyamhpGjx6N2+3mpZdeAuCpp57iBz/4AUOHDsXlcjFx4kT+8pe/fO3zp6amMn/+fIYPH05CQgITJkw4p/oeeeQRbrjhBl555RUmTZpEbGws4eHhF3TOIiIiIt5U63Kzt6iCzIIyMvLLycw7RlD+BsbUrGOGXzrj/Yqox4+s4CGs7n4NNmUWCX36c3d0R4ICmn+GiPHVFoO0tDSbnp7+hWMZGRkMGDDASxWdXmJiIunp6URHR3u1jpqaGvz9/QkICGD9+vXceeedbN682as1XYjW/DkXERGRpmWtpaiihsz8cjLyy8gscN7uLaqA+jrG+u3gcv90ZgRsoJMtod4EUtpjPEGDr6Dj0DkQdn45zBizwVqbdr51a4W7ncnOzua6667D7XYTFBTE3/72N2+XJCIiIvIlNa56dh+uILOgnMxG4bq48uT+JgnhhmuiMpkS+ynJJesIdJVjA8Mw/afDgDn4J02jc0iEF8/CocDdQg4cOODtEgDo168fmzZt8nYZIiIiIoCzan24rIaMgrJGK9dl7C2qpN7tdGIEB/iRHBPOpQO6MbQLjHalk1D4LkH734Oi4xDaCQZfAQNmY/pMgcAQL5/VFylwi4iIiEiLqK5zVq0zCsqcYJ1fTmZBGceq6k7cp2dUKCkx4UwfGENKrDMlJDG4nIDdK50Z2Ws/BLcLwmNhxE0wYA4kXAz+rTfWtt7KRERERMQnWWvJL63+Qp91Rn4Z+49U4lm0JjTQn/4x4cwcHENKTAQpMeGkxEYQGRro3OHYAchYBEuXQs6ngIXOfWDsD2DAXOiRCn6+sWm6AreIiIiInLeqWhe7Dlec6LPemV9GZn4ZZdWuE/eJ6xTKgNgIZg2JJSU2ggGxEcR37oC/X6O9RayFokz4bClkLIECZ0Y23YfA5AedlexuA+As9yNpTRS4RUREROSsHC6rZmtu6Yk+68z8cvYXV9Iw9K5DkD8pMeHMHtaDAbERDIgJp39MOBEhgV/9hNbCoY1OwM5cBsV7nOO9RsP0X0LKbOjcu2VOrhkpcDeDJ554gv/85z/4+/vj5+fHX//6V0aPHt0sr/Xoo4/SsWNH7r33Xh5++GEmTpzI1KlTz+u5/P39GTJkyImP33jjDRITE5uo0vP35JNPctttt53Yzl5ERERazsHiSlZtL2Dl9gI255ScOJ7QpQMpMeHMHd6DlJgIBsSG06tTB/z8zrACXe+C7I+dfuzM5VB2CPwCIHECjPk+pMyC8JjmPakWpsDdxNavX8+yZcvYuHEjwcHBHDlyhNra2jM/sAk8/vjjF/T40NDQ85rJ7XK5CAhovn9KTz75JDfddJMCt4iISAvZfbiclZ6QnZFfBsCQnpHcNyOZMX26kBwTTsfgc/jZX1cN+96HzKWQuQKOH4WAEEiaCpc+DP1nOJNG2igF7iaWn59PdHQ0wcHOlqANG90kJiYyf/581qxZA8B//vMfkpKSKCoq4o477iA7OxtwwuXFF1/Mo48+SnZ2Nvv27SM7O5sFCxZw1113Ac4K+osvvkivXr3o2rUrI0eOBOCWW25h9uzZXHPNNSQmJnLzzTezdOlS6urqePXVV0lJSaGoqIgbb7yR4uJiLrroIlatWsWGDRu+dkOezZs3c8cdd1BVVUXfvn157rnn6NSpE5MnT2bcuHGsW7eOuXPnMnnyZO655x4qKiqIjo7m+eefJzY2lj179nDHHXdQVFSEv78/r776Kt27d+eKK67g2LFj1NXV8ctf/pIrrriCyspKrrvuOnJzc6mvr+ehhx7i8OHD5OXlMWXKFKKjo0/8/YmIiEjTsdayI6/Ms5Kdz96iSoyBkfGd+PmsAcwYFEOvzue48FVTDrtXQ8Yy521tBQRHQP+ZMGC2E7aDwprnhFqZthu4Vz5wstm+qcQMgct+fdq7TJ8+nccff5z+/fszdepU5s+fz6RJkwCIiIjgs88+48UXX2TBggUsW7aMu+++mx//+MeMHz+e7OxsZsyYQUZGBgCZmZmsWbOG8vJykpOTufPOO9m6dSsvv/wymzZtwuVykZqaeiJwnyo6OpqNGzfypz/9id/97nc8++yzPPbYY1xyySU8+OCDrFq1imeeeebE/Y8fP87w4cMB6N27N4sXL+Zb3/oWf/zjH5k0aRIPP/wwjz32GE8++SQAJSUlfPDBB9TV1TFp0iTefPNNunbtyiuvvMLPfvYznnvuOb7xjW/wwAMPMG/ePKqrq09suLN48WIiIiI4cuQIY8aMYe7cuaxatYoePXqwfPlyAEpLS4mMjOT3v/89a9as8founSIiIm2J223ZlFPCWzuckJ1z9Dh+Bsb06cIt4xKZMSiGbhHnOM+6shh2ecb37V0D9TUQ1hUGX+1MFuk9EQKCmueEWrG2G7i9pGPHjmzYsIG1a9eyZs0a5s+fz69/7YT0G2644cTbH//4xwC888477Ny588Tjy8rKKC8vB2DWrFkEBwcTHBxMt27dOHz4MGvXrmXevHkn2ivmzp37tbVcddVVAIwcOZJFixYB8NFHH7F48WIAZs6cSadOJ399c2pLSWlpKSUlJSf+w3DzzTdz7bXXnrh9/vz5AGRlZbF9+3amTZsGQH19PbGxsZSXl3Po0CHmzZsHQEiI80VbV1fHT3/6Uz788EP8/Pw4dOgQhw8fZsiQIdx7773cf//9zJ49mwkTJpzl37qIiIicjXq35bP9R1m1PZ+3dhymoKyaQH/DxUnR/HBKEtMGxtA57BwDcekhpxc7YwkcXAfWDZG94KLvOJNFeo0GP//mOSEf0XYD9xlWopuTv78/kydPZvLkyQwZMoQXXngBANNojE3D+263m/Xr1xMaGvql52loS2l4TpfL9aXnOZ2Gxzd+rG24jLgJhIWFnXjOQYMGsX79+i/cXlZW9pWP+/e//01RUREbNmwgMDCQxMREqqur6d+/Pxs2bGDFihU8+OCDTJ8+nYcffrjJ6hUREWmPal1u1u8rZtX2fFbvOExxZS3BAX5M6t+V+4ckc0lK95Ozr89W8V4nYGcshUMbnGPRyTD+Hidkxw7zyfF9zcU3poX7kKysLHbv3n3i482bN5OQkADAK6+8cuLt2LFjAacF5emnn/7C/U9n4sSJLF68mOPHj1NeXs7SpUvPqb7x48ezcOFCAFavXs2xY8e+9r6RkZF06tSJtWvXAvDPf/7zxGp3Y8nJyRQVFZ0I3HV1dezYsYOIiAji4uJ44403AKipqaGqqorS0lK6detGYGAga9as4eDBgwDk5eXRoUMHbrrpJu699142btwIQHh4+IlVfxERETmz6rp6Vu8o4J6Fm0n75dvc/NxnLNmcx7ikaP70jVQ2PjSNZ76VxrwRcWcXtq2F/C3w3hPwp7Hwx1R451FnNfvSh+EHn8MPP4NLH4IewxW2T9F2V7i9pKKigh/96EeUlJQQEBBAUlISzzzzDMuWLaOmpobRo0fjdrt56aWXAHjqqaf4wQ9+wNChQ3G5XEycOJG//OUvX/v8qampzJ8/n+HDh5OQkHDObRePPPIIN9xwA6+88gqTJk0iNjaW8PDwr73/Cy+8cOKiyT59+vCPf/zjS/cJCgritdde46677qK0tBSXy8WCBQsYNGgQ//znP7n99tt5+OGHCQwM5NVXX+Ub3/gGc+bMIS0tjeHDh5OSkgLAtm3buO+++/Dz8yMwMJA///nPANx2221cdtllxMbG6qJJERGRr1FZ42JNViErtxewJrOQqtp6IkMDmTYwhssGxzC+XzQhgefQ2uGud3Z4zFjmTBcpyQbjB/FjYeZvnPF9Ub2a74TaENOULQYtKS0tzaanp3/hWEZGBgMGDPBSRaeXmJhIenq61y/8q6mpwd/fn4CAANavX8+dd955XqMAW4vW/DkXERFpbqXH63g34zArtxfw4a4ialxuojsGnQjZY/t2IdD/HBoaXDWw7wMnYGethMoi8A+CPlOcySLJl0NY+xtiYIzZYK1NO9/Ha4W7ncnOzua66647MS3kb3/7m7dLEhERkXNQXFHD6p2HWbW9gI/3HqGu3hIbGcINo+K5bHAMaYmdv7hl+pnUlMPut52dHnethtpyCAqHftM84/umQUhE851QO6DA3UIOHDjg7RIA6NevH5s2bfJ2GSIiInIOCkqrT4zv+2z/UdwW4jt34NaLezNzcAzD4qLOvMNjY5XFkLXCCdkN4/s6RMPgeZAyB/pMgoDgMz+PnJU2F7ittWc9xUN8m6+2Q4mIiJyNnKNVJzai2ZhdAkBSt478YEoSMwfHMDA24twyT0mOE7Azljlbq1s3RMY74/tSZkP8mHY/vq+5tKnAHRISQnFxMV26dFHobuOstRQXF5+Y7S0iItIW7CmsYNX2fFbtKGD7IWe87qAeEdw7vT8zB8eQ1O3rBx18ibVQlOX0Y2csg/zNzvGuA2DCT5zxfTFDNVGkBbSpwB0XF0dubi5FRUXeLkVaQEhICHFxcd4uQ0RE5LxZa8nIL2fV9nxWbi9gd2EFACPio/jp5SnMHBRLfJdz2FLd7Ya8TZ6QvRSK9zjH4y6CqY85IbtL32Y4EzmdNhW4AwMD6d27t7fLEBEREfla1lq25Jaycns+q7YXcLC4Cj8DFyV25tE5A5kxOIbYyC9viPe16uucHR4zljk7PpbngV8AJI6H0Xc44/siejTfCckZtanALSIiItIa1bst6QeOsnJ7AW/tKCC/tJoAP8O4pGjumNSXaQO7E93xHC5SrDsOe99zQvaulXD8GASEQtKlMOAR6D8DQjs13wnJOVHgFhEREWkGdfVuPtlXzMrtBazecZgjFTUEBfgxsV9X7p2ezNQB3YnscA5bqh8vgV1vOe0ie96FuioIiYT+lznj+/peCkHn0H4iLUaBW0RERKSJ1Ljq+Wj3EVZuL+CdjMOUVNXRIcifKcndmDk4hikp3egYfA7xq7zAaRPJXAb7PwS3CzrGwLAbnJCdOAH8zyG0i1cocIuIiIhcgKpaF+9nFZ3YUr2ixkV4SADTBnRnxuAYJvXvem5bqh/d57SKZCyF3M8BC537wJjvw4C50HMk+J3D7pHidQrcIiIiIufA7bbsO1LJxoPHeDfzMB/sKqK6zk3nsCBmD41l5uAYxvWNJijgLEOxtVCw7eSM7MIdzvGYoTDlp86M7G4DNL7Phylwi4iIiJxGeXUdm3NK2HiwhI3Zx9icU0Lp8ToAuoUHc11aL2YOjmFUYmcC/M8yZLvrIeczT8heCiUHAQPxY2HGr5yQ3Smh+U5KWpQCt4iIiIiHs3pdwcaDJWzKOcbGgyXsKizHWmeBuV+3jlw2OIbU+E6MiI+ib9eOZ7+luqvW6cPOXOr0ZVcWgX8Q9JnsbESTfDl07Nqs5yfeocAtIiIi7VZZdR2bs0vYlO2sXm/KPkZZtQuAiJAARsR34vIhsaQmRDGsVxQRIed4gWJNBex522kV2b0aasogqCP0m+asYvebDiERzXBm0poocIuIiEi74HZb9hZVnAjXG7OPsbuw4sTqdf9u4cwaGsuI+E6kxneiT3TY2a9eN1ZZ7MzGzljmzMqur4EOXWDgXOeix96TIDCk6U9QWi0FbhEREWmTSo/XsSWnIVyXsLnR6nVkaCAj4qOYPbQHqfGdGNor8txXr7/wYrlOm0jGUmfXR+uGyF6Qdqszvq/XGPBX7Gqv9JkXERERn9ewer0x+9iJixv3FJ1cvU7uHs6soT1IjY9ixIWsXjdWtAsyljgXPuZtco51TYHx9zghO3a4JosIcBaB2xjzHDAbKLTWDvYc+y0wB6gF9gLfttaWNHpMPLATeNRa+zvPsZHA80AosAK421prjTHBwIvASKAYmG+tPdBE5yciIiJtUOnxhskhx05MDin3rF5HdQhkRK8o5g7rwYj4TgzrFUn4haxeN3C7nWCdudRpFyne7RzvORKmPgopcyA66cJfR9qcs1nhfh54GicUN3gbeNBa6zLG/AZ4ELi/0e1/AFae8jx/Bm4DPsEJ3DM99/kOcMxam2SMuR74DTD/3E9FRERE2iK327KnqOJEuN6UXcLuwgoA/Az07x7OnGE9GNEritQEZ/XaNNXKcn0dHPjIWcXOXAHleeAXAInjYfTtzmSRyJ5N81rSZp0xcFtrPzTGJJ5ybHWjDz8Brmn4wBhzJbAPqGx0LBaIsNau93z8InAlTuC+AnjUc9fXgKeNMcZaa8/5bERERMTnlVbVOSP5skvY9BWr16nxnbhieMPqddS5bZV+NmorYc+7TsjetQqqSyEgFJIuhQGPQP8ZENqpaV9T2rSm+Bd6K/AKgDEmDGelexpwb6P79ARyG32c6znWcFsOgGfFvBToAhxpgtpERESkFXO7LbsLK06M5NuYXcKeRqvXyTERJ1pDUuOj6N2Uq9eNVR11wnXDZBHXcSdUJ89y+rH7TIGgDk3/utIuXFDgNsb8DHAB//Ycegz4g7W24pQvhq/6yrBncdupr3cbTlsK8fHx51OyiIiIeFFpVR0bc5y2kE3Zx9icXUJ5jbN63cmzen3l8IbJIc2wev2FYhpPFvkYbD1E9ITUb0HKLEi4WJNFpEmc978iY8zNOBdTXtqo/WM0cI0x5n+AKMBtjKkGXgfiGj08DsjzvJ8L9AJyjTEBQCRw9Kte01r7DPAMQFpamlpOREREWrF6t2V3Ybkz99rTf723yOk49TOQEhPBXE+4Tk3oRGKXDs2zet3AWijKOnnRY/5m53h0Moxf4GxE02OEJotIkzuvwG2MmYnTOjLJWlvVcNxaO6HRfR4FKqy1T3s+LjfGjAE+Bb4F/NFz1yXAzcB6nF7w99S/LSIivqqovIYlW/LYfqgUY8DfGPz9DH5+5sT7X3X85O184b5+je7n53ncF25v/LizfK3Gz/uFx504dubH+Rm+FI5LqmrZlFPCpoOeudc5JVR4Vq87hwWRGh/FValxjIiPYlhcFGHNuXrdwO2GQxtObqdevMc53jPNM1lkNkT3a/46pF07m7GALwGTgWhjTC7wCM5UkmDgbc8X2yfW2jvO8FR3cnIs4EpOTjH5O/BPY8wenJXt68/5LERERLyouq6edzIOs2jjIT7YVUS929IjMgQ/P4Pbbam3lno3uK2l3m0bHbMnj/ngUpOf4QshvKq2HnCOpcSEc+UIz+p1fCcSmnv1ujFXLRxY6wTsrBVQnu+ZLDIBxtzpTBaJ6NEytYgAxlcXk9PS0mx6erq3yxARkXbKWkv6wWMs2pjLsq35lFe7iI0MYd6InlyV2pOkbuHn/Hxuy4kQfjKc86Vw/sWg7oT5Lz3ulOP1npDvdp9yu+fYlx73pfs2fnyj2z2v6XZbojoEOb3XcZEts3rdWG0l7HnHaRXZ9RbUlEJgB2eySMoc6D9dk0XkvBljNlhr08738boSQERE5BwcLK5k0cZDLN50iOyjVXQI8ueywbFcndqTMX26nPfuhcYY/D0rxnKWKoth10onZO9bA65qCO3sTBVJmQ19p0BgqLerFFHgFhEROZPS43Us35rPoo25pB88hjEwPimaH0/rx4xBMXQI0o/TFlOS7ZkssgyyPwbrhog4GHmLE7Ljx2qyiLQ6+hcpIiLyFerq3azdXcTrGw/x9s7D1LrcJHXryP0zU7hyRA9iI7Vy2iKshcIMJ2RnLoX8Lc7xrgNg/D3OanbscE0WkVZNgVtERMTDWsuOvDIWbTzEki2HOFJRS+ewIG4cFc/VqXEM7hnRchf+tWduNxxKd+ZjZy6Do/uc43GjYOpjMGAOdOnr3RpFzoECt4iItHuHy6p5Y9MhFm08RNbhcoL8/Zg6sBtXjYhjUnJXAv39vF1i2+eqhQMfOq0iWSug4rAzWaT3RBj7Q2cjmvAYb1cpcl4UuEVEpF2qqnWxesdhXt+Yy7o9R3BbSI2P4pdXDmb20FiiOgR5u8S2r6YC9rzthOzdq6GmDALDoN9UZ7JIv2kQGuXtKkUumAK3iIi0G2635dP9R1m0MZcV2/KprK0nrlMoP5ySxLzUOHpHh3m7xLav8oizgp25HPaugfoa6NAFBs51LnrsM1mTRaTNUeAWEZE2b29RBYs9o/wOlRynY3AAs4f24KrUnlyU2Pm8R/nJWTp20HPR4zLIXu9MFonsBWm3Ohc99hqjySLSpulft4iItEnHKmtZtjWP1zceYnNOCX4GJvTryn/NTGb6wBhCg/y9XWLbZS0U7nRaRTKXQsE253i3gTDhXidkxwzVZBFpNxS4RUSkzah1uVmTVciijbm8l1lIXb0lJSacn10+gCuG96BbRIi3S2y73G7I/cwzWWQ5HNsPGOg1Cqb9wrnoUZNFpJ1S4BYREZ9mrWVrbimLNuayZEsex6rqiO4YzM1jE7kqNY6BPSK8XWLbVXcc9r3vBOxdq6CyCPwCnckiF98FybMgvLu3qxTxOgVuERHxSXklx1m86RCLNuayt6iS4AA/pg+K4arUnkxIiiZAo/yaR+URJ1xnroC974HrOASFOxNFUmY5b0MivV2lSKuiwC0iIj6josbFqu0FLNqYy/p9xVgLoxI7870Jfbh8aCwRIYHeLrFtKt7rrGJnrYCcTz3bqfeEEd+A5MshcQIEaIyiyNdR4BYRkVat3m35eO8RFm08xKrtBRyvqyehSwcWXNqfeSN6Et+lg7dLbHsadnrMXA5ZK+FIlnO8+xDnoseUy7Wdusg5UOAWEZFWadfhchZtPMQbmw5RUFZNREgA81J7cnVqT1LjO2mL9aZWdxz2fQBZyyFrFVQWgvGHxIud8X3Jl0GnBG9XKeKTFLhFRKTVKK6oYcmWPBZtPMS2Q6X4+xkm9+/KQ7MHcumAboQEapRfk6osdvqxszz92HVVnn7sqU6rSL9pENrJ21WK+DwFbhER8arqunrey3RG+b2fVYTLbRncM4KHZw9k7vAeRHcM9naJbUvxXs9Ojysg5xOnHzu8Bwy7wWkVSZwAAfo7F2lKCtwiItLirLVszD7G6xsPsWxLHmXVLrpHBPOdCb25akQcyTHh3i6x7XC7IW/jyYseizKd490Hw4SfOCvZPUaoH1ukGSlwi4hIi8k5WnVilN+B4ipCA/2ZOdgZ5TeubzT+2mK9adRVw/4PTs7Hrjjs9GMnjIORt3j6sRO9XaVIu6HALSIizaqyxnVii/XP9h8FYGyfLvxgShKXDYmlY7B+FDWJqqOw6y3nosc970FdJQR1hKRLnQ1o+k2DDp29XaVIu6TvciIi0iwKSqt5/uMD/OfTg5RVu+gTHcZ9M5K5YngP4jpplF+TOLrP6cXOWgHZ6z392LEwbL4TsnurH1ukNVDgFhGRJrUzr4xn1+5jyZY83NYyc3AMt17cm5EJGuV3wRr6sRsueizKcI53Gwjj7/HMxx4BftplU6Q1UeAWEZEL5nZbPthVxLMf7WPdnmI6BPlz05gEbr24tzamuVB11bD/w5PzsSsKTvZjp/6304/dube3qxSR01DgFhGR81ZdV88bmw7x7Ef72VNYQfeIYO6fmcKNo+KJ7KBt1s9b1VHYvdq56HHPu04/dmCY04+dMgv6TVc/togPUeAWEZFzVlxRw78+yeafnxzgSEUtA2Mj+MP8Ycwa0oOgALUznJej+0+2imSvB1sPHWNg6HXO6L7eEyEwxNtVish5UOAWEZGztreogr9/tJ/XN+RS43IzJbkr35vQh7F9u6g/+1y53ZC/6eRFj4U7neNdB8D4Bc5Fjz3Ujy3SFihwi4jIaVlr+WTfUZ5du493MwsJCvDj6tSefGd8b5K6aYOac+KqcfqxG+Zjl+eD8YP4cTDjV55+7D7erlJEmpgCt4iIfKW6ejcrtuXzt7X72H6ojM5hQdx9aT++OTZB262fi6qjsPttz3zsd6G2wtOPfYmzit1/hvqxRdo4BW4REfmCsuo6Xvo0m+c/PkB+aTV9uobxq3lDuCq1JyGB/t4uzzccO+jpx14OBz/29GN3hyHXeOZjqx9bpD1R4BYREcDZdv0f6w7wyufZVNbWM7ZPF56YN5jJ/bvhpy3Xz6woCzKWQMZSyN/iHOuaAhff7UwW6ZGqfmyRdkqBW0SknducU8Lf1u5j5bZ8/Ixh9tBYvjuhD4N7Rnq7tNbNWsjb5ATsjKVQvNs5HncRTHscUmZDl77erVFEWgUFbhGRdqjebXl752H+/tE+Pj9wjPCQAL43oQ+3XJxIbGSot8trvdz1zsi+jGVOyC7LdTahSRwPo293VrIjeni7ShFpZRS4RUTakapaF69tyOW5j/ZzoLiKuE6hPDx7INdd1IuOwfqR8JUaJotkLHFG+FUdAf9g6HsJTPmpM1lEFz2KyGnou6uISDtQWFbNC+sP8O9PsympqmN4ryj+b0YKMwZ1J8BffcVfUlMBe95xVrF3r4aaMgjq6EwUGTAHkqZCsEYiisjZUeAWEWnDMvLLeHbtfpZsOYTLbZkxMIbvTezNyAStyH5J1VHY9ZYTsve+C65qCO0MA+fCgLnQe5Imi4jIeVHgFhFpY6y1fLj7CM+u3cfa3UcIDfTnxlHx3Dq+NwldwrxdXutSXgCZnn7sAx+B2wXhPSD1ZmclO34s+OtHpYhcGH0XERFpI2pc9by5OY+/r91P1uFyuoUH818zk7lxVDxRHYK8XV7rcXT/yZCd8xlgoXNfGPcjSJmj7dRFpMkpcIuI+LhjlbX865ODvLD+IEcqakiJCed/rx3GnGE9CApQcMRaKMw4Ob7v8DbneMwQ56LHAXOcedlGs8ZFpHkocIuI+Kj9Ryr5+0f7eG1DLtV1bib178r3JvTh4qQumPYeHt1uyNt4MmQf3QsY6DUapj8BA2ZDp0RvVyki7YQCt4iID7HW8vmBY/xt7T7eyThMoJ8f80b05DsTetO/ezufmlHvguyPPSF7GZTngV8AJE6AsT9wZmSHx3i7ShFphxS4RUR8gKvezYrtBTy7dh9bc0vp1CGQH01J4ptjE+kaHuzt8rynrhr2ve+E7KwVcPwoBIQ4Y/tSHnbG+GlGtoh4mQK3iEgrVl5dxyuf5/CPdQc4VHKcPtFhPDFvMFeNiCM0yN/b5XlHTTnsfvvkjOzaCgiO+OKM7CBNYxGR1kOBW0SkFTpUcpzn1+3n5c9yKK9xMbp3Zx6bO4hLUrrh59cO+7OrjkLWSs+M7PegvgY6RMPgqz0zsidAQDte6ReRVk2BW0SkFdmaW8Lf1u5nxbZ8AGYNieV7E/owJC7Sy5V5QVkeZC53tlQ/sA5sPUTEQdqtnhnZY8Cvna7yi4hPUeAWEfEyt9vybmYhf1u7j8/2HyU8OIDvjO/NzeMS6RkV6u3yWlbxXmcVO3MZ5H7uHOvSD8YvcEJ27HCN7xMRn6PALSLiJcdr63l9Yy7PfbSffUcq6RkVys9nDWD+Rb0IDwn0dnktw1o4vOPk+L7CHc7x2OFwyc+ddpGuyV4tUUTkQilwi4i0sMLyav65/iD/+uQgx6rqGNYriqdvHMHMQTEE+LeDjWrcbjiU7rSKZCyDY/sB42yjPuO/nfF9nRK8XaWISJNR4BYRaQH1bsu6PUdYmJ7D6h2HqXO7mTagO9+b2Ie0hE5tf6Oa+jo4uO7kjOyKAvALhD6T4OK7nZDdsZu3qxQRaRYK3CIizSi7uIrXNuTw2oZc8kqrieoQyI2j47l5XCK9o9v46Lp6Fxz4ELa/7lz8ePwYBIRCv6lOq0i/6RAa5e0qRUSa3RkDtzHmOWA2UGitHew59ltgDlAL7AW+ba0tMcaMAp5peCjwqLV2secxI4HngVBgBXC3tdYaY4KBF4GRQDEw31p7oMnOUESkhR2vrWfVjnwWfp7L+n3FGAMT+3XlZ7MGMnVgN4ID2vBkDbcbstc7IXvnm1B1BILCIfkyGHgF9L0Egjp4u0oRkRZ1NivczwNP44TiBm8DD1prXcaY3wAPAvcD24E0z/FYYIsxZqm11gX8GbgN+AQncM8EVgLfAY5Za5OMMdcDvwHmN8nZiYi0EGstW3JLWZiew9LNeZTXuIjv3IF7p/fnqtQ4erTlaSPWwqENTsjesRjK852V7OSZMOgq6DcNAtvw+YuInMEZA7e19kNjTOIpx1Y3+vAT4BrP8apGx0MAC+AJ3xHW2vWej18ErsQJ3FcAj3oe8xrwtDHGWGvtuZ+OiEjLOlJRwxubDrEwPYddhysICfTj8iGxXJfWi1GJndvuJjXWQsFW2L4IdiyCkmzwD4KkaTD4Kug/E4I7ertKEZFWoSl6uG8FXmn4wBgzGngOSAC+6Vnt7gnkNnpMLtDT835PIAfAc99SoAtwpAlqExFpcq56Nx/sKmJheg7vZhTicltGxEfx31cNYfbQ2LY90q8w07OSvQiK94Dxh75TYPKDkHy5erJFRL7CBQVuY8zPABfw74Zj1tpPgUHGmAHAC8aYlTj93KdqWME+3W2nvt5tOG0pxMfHX0DlIiLnbm9RBa+m57JoYy6F5TVEdwzi1vG9uXZkHP26h3u7vOZTvNcJ2NsXe+ZkG2cr9bE/dC5+DOvi7QpFRFq18w7cxpibcS6mvPSr2j+stRnGmEpgMM6Kdlyjm+OAPM/7uUAvINcYEwBEAke/6jWttc/guSgzLS1NLSci0uwqalys2JrPwvQc0g8ew9/PMCW5G9elxTElpRuBbXVudkmO04+9/XXI3+wc6zUGLvsf5+LH8Bivlici4kvOK3AbY2biXCQ5qXHftjGmN5DjaQ1JAJKBA9baI8aYcmPMGOBT4FvAHz0PWwLcDKzH6QV/T/3bIuJN1lrSDx5j4ec5LN+WT1VtPX27hvHgZSnMS+1Jt/AQb5fYPMoPw843nJCd86lzrMcImP5LGHglRPXyZnUiIj7rbMYCvgRMBqKNMbnAIzhTSYKBtz2bNXxirb0DGA88YIypA9zA9621Db3Yd3JyLOBKzx+AvwP/NMbswVnZvr5JzkxE5BwdLqvm9Y25vJqey/4jlYQF+TN3WA+uTetFanxU29ycprIYMt50Ln488BFgoftguOQh5+LHzn28XaGIiM8zvrqYnJaWZtPT071dhoj4uFqXm/cyD7MwPZf3swpxWxjVuzPXpfXi8iExdAhqg/uDHS9xNqLZsQj2rgFbD12SYPDVzhi/binerlBEpFUxxmyw1qad7+Pb4E8SEZEzyyooZ2F6Dos3HeJoZS3dI4K5c3JfrhnZq23uAFlTAbtWOe0ie96B+lqIioeL73JCdswQaIsr+CIirYACt4i0G6XH61i6JY9X03PYkltKoL9h2sDuXJvWi4n9uuLf1mZm1x2H3W87IXvXW+A6DuGxcNH3nNXsnqkK2SIiLUCBW0TaNLfb8sm+Yham57ByewE1LjcpMeE8PHsgV47oSeewIG+X2LRctbD3PaddJHM51FZAWFcY8Q0nZPcaA35tdLKKiEgrpcAtIm3SoZLjvJaey6sbcsg9dpzwkACuS+vFdWm9GNwzom1dAFnvggMfOhc+ZiyF6hIIiXIuehx0FSROAH99uxcR8RZ9BxaRNqO6rp7VOw/zanoOH+05grUwPima+2YkM2NQDCGB/t4usem43ZC93mkX2fkmVB2BoHBImeUE7T5TIKCNrd6LiPgoBW4R8XnbD5WyMD2HNzYdoqzaRc+oUO6+tB9Xp8bRq3MHb5fXdKyFQxs8W6u/AeV5EBAKyTOddpGkqRAY6u0qRUTkFArcIuKTjlXW8sbmQyxMzyUjv4ygAD8uGxzDdWm9GNunC35t5QJIa6Fgq9MusmMRlGSDfxAkTYPBv4D+MyG4o7erFBGR01DgFhGfUe+2rN1dxKvpuby98zC19W6GxkXyiysHM3doDyI7BHq7xKZTmOkE7O2vQ/Ee8Atw2kQmP+i0jYREertCERE5SwrcItLqHSyu5NX0XF7fmEt+aTWdOgRy05gErk2LY0BshLfLazrFez0hezEU7gDjB4njYdyPYMBc6NDZ2xWKiMh5UOAWkVapqtbFym0FLEzP4dP9R/EzMLF/Vx6ePZBLBnQjOKCNXABZkgM7FjtBO2+Tc6zXGLjstzDwCgjv7t36RETkgilwi0irYa1lU04Jr6bnsHRLPhU1LhK6dOC+GclcldqT2Mg2ckFg+WHY+YbTLpLzqXOsRypM/yUMmgeRcV4tT0REmpYCt4h4XVF5DYs35bIwPZc9hRWEBvoza2gs146MY1Tvzr4/M9taOLIbslZA1kpPyLbQfTBc8pAzxq9zH29XKSIizUSBW0S8JudoFb9cvpN3MwpxuS2p8VH8+qohzB7Wg47BPv7tqd7lBOuGkH10r3M8dhhMfsBZye6a7N0aRUSkRfj4TzQR8UXWWhZtPMQjS3ZggO+M7821aXEkdQv3dmkXpqYc9rzrBOzdb8HxY84Iv94TYez3of9lENnT21WKiEgLU+AWkRZVWlXHT9/YxvKt+YxK7Mzv5w8jrpMPb05Tegh2rXRC9v4Pob4WQjs587GTL4O+l0Cwj/9HQkRELogCt4i0mI/3HuEnC7dQVF7DfTOSuWNSX/x9bYOaho1oslY67SL5W5zjnfvAqNucGdlxo8Bf315FRMShnwgi0uxqXPX8fvUunlm7j95dwlj0/XEMjYvydllnz1UDBz7y9GOvgrJcwECv0TD1MUi+HKL7ga9f3CkiIs1CgVtEmtWewnLuemkzO/PLuGFUPA/NHkCHIB/41lN1FHa/7YTsPe9CbTkEdnBaRKb8FPpNh45dvV2liIj4AB/4qScivshay78+Ocgvl2cQFhzAM98cyfRBMd4u6/SK93paRVZC9nqw9dAxBoZc7axi954IgW1kFriIiLQYBW4RaXJF5TXc//pW3sssZFL/rvz22qF0Cw/xdllf5q6HQxucVezMFXAkyznefTBMuMe56DF2BPj5ebdOERHxaQrcItKk3s04zH+9tpXyGhePzhnIzeMSW9fGNbWVsO99J2Tvegsqi8AvABIuhrRbIXkmdEr0dpUiItKGKHCLSJM4XlvPEyt28q9PskmJCec/3xtDckwrGYdXXgC7VjmtIvveB1c1BEdCv2nOKnbSVAiN8naVIiLSRilwi8gF236olLtf3sTeokq+O743981MJjjA33sFWQuFGSd3eTyU7hyPioeR33ZCdsI48A/0Xo0iItJuKHCLyHmrd1v+tnYf/7s6i85hQfzrO6MZ3y/aS8XUwcGPT87HLjnoHO85Ei75uXPRY7eBGt0nIiItToFbRM5LXslx7lm4mU/2HeWywTH8at4QOoUFtWwR1aWe0X0rnbc1pRAQAn0mOxc99p8J4a18MoqIiLR5Ctwics6WbsnjZ4u3Ue+2/M81Q7l2ZFzLXRh57KCnH3uFsxmN2wUdomHAHEi53AnbQWEtU4uIiMhZUOAWkbNWXl3HI2/uYNGmQwzvFcX/u344CV2aOdy63ZC/6eR87MPbnePRyTD2h06rSFwa+HmxZ1xEROQ0FLhF5KykHzjKglc2k1dynLsu7cePLkki0L+Z5lPXVcP+DyFrubOVekUBGD+IHwfTn3AueuzSt3leW0REpIkpcIvIadXVu3nq3d3835o99OwUyqt3jGVkQuemf6HKI85c7KwVsPc9qKuCoI6QdKmzit1vOnRohtcVERFpZgrcIvK19h+pZMErm9mSU8LVqXE8Oncg4SFNNErPWjiy62SrSM6ngIWInjD8RmcVO3ECBAQ3zeuJiIh4iQK3iHyJtZaF6Tk8tnQngf5+/N+NqcwaGnvhT1xfB9nrnTaRrBVwbL9zPGYoTH7ACdkxQzW6T0RE2hQFbhH5gmOVtTywaCtv7TjMuL5d+N/rhhEbGXr+T3j8GOx+B3athD3vOKP8/IOh90QY90NndF9kXNOdgIiISCujwC0iJ6zdXcRPFm7hWFUtP708he+O74Of33msNhfvddpEdq1yNqOx9RDWFVLmOKvYfSZDcMcmr19ERKQ1UuAWEarr6vmfVVk8t24/Sd068twtFzG4Z+TZP0G9y+nB3rXSaRcp3u0c7zYIxi+A/pc5Oz76NdNUExERkVZMgVukncssKGPBy5vJLCjnW2MTePCyAYQGncVM6+pS2POus4q9e7XTOuIXCInjYdT3nFaRTgnNfwIiIiKtnAK3SDvldlue//gAv16VSURIAP+45SKmpHQ7/YOO7vfs8rgSDq5zdnkM7eyE6/4zoe8lEBLRMicgIiLiIxS4RdqhwrJqfvLqFtbuPsKlKd34zTVDie74FeP33PWQm36yVaQowzkenQxjf+C0ivQapV0eRURETkOBW6SdeWtHAQ+8vpXjdfX88srBfGN0PKbxGL6acmfjmaxVsPstqCoGvwCIHwup/w3JM6FzH++dgIiIiI9R4BZpJyprXPxi2U5e/jyHwT0jeHL+CJK6eSaFlOScbBU5sBbqayEkCvpNc1pFkqZCaJQ3yxcREfFZCtwi7cDmnBIWvLyJg0eruGNSX+6ZmkTQ4c3wrmd03+Htzh0794VRtzmj+3qNAX99ixAREblQ+mkq0obVuy1/fn8Pf3hnNwkdLStnlJNS+jQ8+RZUFoLxc1pFpv3CCdnR/bxdsoiISJujwC3SRuUcreKXL71NdN77LOm0k4HVmzDv10BwBCRd6lzw2G8adOjs7VJFRETaNAVukbbEWmzeZjI/XIjNXMlfzX4IBBuUiBl8q3PBY/w4CAjydqUiIiLthgK3iK+rOw77PoBdK3FnrcKvooBka9gVNICSi35K1PArMF2TwZzHFu0iIiJywRS4RXxR+WHnYsddq2DvGnAdpz4gjA/dQ1jhupL+F1/Nt6enEeCvrdRFRES8TYFbxBdYCwXbTo7uy9voHI/sRf3wb/B6xRAe2hJFbOdInvz2CIb3ivJquSIiInKSArdIa+Wqgf1rT+7yWJbrHO+ZBpf8HPpfxh6TwIKFm9l+qIzrL+rFQ7MHEhasL2sREZHWRD+ZRVqTiiJnd8eslU6rSF0lBHaAPlNg8v3QbwaEd8day78/zeaXyz8iNNCfv9w0kpmDY7xdvYiIiHwFBW4RbyvLg4ylsPNNyF4P1g3hPWDodZB8OfSeAIGhJ+5+pKKGB17fyjsZhUzoF83vrh1G94gQL56AiIiInM4ZA7cx5jlgNlBorR3sOfZbYA5QC+wFvm2tLTHGTAN+DQR5brvPWvue5zEjgeeBUGAFcLe11hpjgoEXgZFAMTDfWnugKU9SpNU5dgB2LoGMJZD7uXOsawpMuBdSZkHssK+cKrIms5D7XttCWbWLh2YP5NvjEvHz0/QRERGR1uxsVrifB57GCcUN3gYetNa6jDG/AR4E7geOAHOstXnGmMHAW0BPz2P+DNwGfIITuGcCK4HvAMestUnGmOuB3wDzL/TERFqdol2Q8aYTtAu2Osdihjr92AOugK79v/ah1XX1/GpFBi+uP0hy93D+9d3RpMREtFDhIiIiciHOGLittR8aYxJPOba60YefANd4jm9qdHwHEOJZwe4MRFhr1wMYY14ErsQJ3FcAj3oe8xrwtDHGWGvteZyPSOthLRze4axi73wTijKd43EXOVupD5gDnXuf8Wl25JVy98ub2VNYwa0X9+a/ZiYTEujfzMWLiIhIU2mKHu5bgVe+4vjVwCZrbY0xpieQ2+i2XE6ufPcEcgA8K+alQBec1XIR32KtM7KvoV3k6D7AQMI4mPkbJ2RH9jzj0wC43ZZnP9rHb9/KolOHIF68dRQT+3dt3vpFRESkyV1Q4DbG/AxwAf8+5fggnNaQ6Q2HvuLh9ixuO/X1bsNpSyE+Pv48KhZpBm435H7mrGJnLIXSHDD+0HsijPsRpMyGjt3O6SnzS4/zk4Vb+HhvMTMGdee/rxpK5zBtxy4iIuKLzjtwG2NuxrmY8tLG7R/GmDhgMfAta+1ez+FcIK7Rw+OAvEa39QJyjTEBQCRw9Kte01r7DPAMQFpamlpOxHvqXXBwnbOKnbEMKgrAPwj6XgKTH4Tky6BD5/N66uVb8/np4m3Uutz85uohXJfWC6Nt2UVERHzWeQVuY8xMnIskJ1lrqxodjwKW41xQua7huLU23xhTbowZA3wKfAv4o+fmJcDNwHqcXvD31L8trZKrFvZ/4KxkZ62AqmIICIV+U52LHvvPgJDzu5DR7bZsyjnGi+sP8ubmPIbFRfLk9SPoHR3WxCchIiIiLe1sxgK+BEwGoo0xucAjOFNJgoG3PStvn1hr7wB+CCQBDxljHvI8xXRrbSFwJyfHAq70/AH4O/BPY8wenJXt65vkzESaQt1x2PueJ2SvgppSCAp3wvXAuZA0FYLOLxRba9mcU8Lyrfks35ZPfmk1QQF+/OiSJO66tB+B/n5NfDIiIiLiDcZXF5PT0tJsenq6t8uQtqimAnavdtpFdq12dnsMiXLmYw+YC30mQ+D5bTRjrWX7oTKWbc1j2dZ8DpUcJ9DfMKl/V2YNjWXqgO6EhwQ26emIiIjIhTHGbLDWpp3v47XTpAjA8RLYtcqZLrL3XXBVQ1hXGHotDLwCEieA//kFYWstGfnlLNuax/Jt+RwsriLAzzC+XzQ/ntafaQO7ExmqkC0iItJWKXBL+1VZDFnLnXaRfR+Au87ZUj31ZqddJH4s+J3/vOusgnKWe1ay9x2pxN/PMK5vF74/uS8zBsUQ1UFTR0RERNoDBW5pX8oLnNF9GUvgwDqw9RCVAGPucC587DkS/M6/d3pPYYWzkr01n92FFfgZGNOnC9+Z0JuZg2Lo0jG4CU9GREREfIECt7R9JTme3R6XQM6ngIUu/WD8AqcnO3YYXMDYvf1HKk+sZGcWlGMMXJTYmV9cMYgZg2PoFn5+/d4iIiLSNihwS9tUvPdkyM7b6BzrPtiZkT1wLnRNuaCQnV1cxfJt+SzbmseOvDIA0hI68cicgVw+JJbuEQrZIiIi4lDglrbBWijKPLml+uHtzvEeqTD1UWclu0vfC3qJQyXHWe5pF9mSWwrA8F5R/HzWAC4fEkuPqNALPAkRERFpixS4xXdZCwVbnZC9800o3g0Y6DUaZvwKBsyBqPgLeomC0uoTK9mbsksAGNIzkgcvS+HyIbH06tzhws9DRERE2jQFbvEtbjcc2gAZbzpBu+QgGD9IHA+jb3dCdnjMBb1EYVk1K7cXsGxrHp8fOAbAwNgI7puRzOyhsSR00e6PIiIicvYUuKX1c9dD9npPu8hSKM8Dv0BnA5qJ90LyLAjrckEvcaSihpXbC1i+NY9P9x/FWkjuHs5PpvXn8qGx9O3asWnORURERNodBW5pneqOw8F1kLEMMpdBZREEhEDfS2HgI9B/JoRGXdBLHKusZdWOApZvzefjvUdwW+jbNYy7LunH7KGx9Ose3jTnIiIiIu2aAre0Dm43FO6Ave85fw6uh/oaCAyD/tOdix77TYfgC1tpLq2q462dBSzbms+6PUeod1sSu3Tg+5OTmD0sluTu4ZgLmF4iIiIicioFbvGe8gLYu8YJ2Pveh8pC53i3gXDRd6HvJZB4MQRe2PSPsuo63tl5mGVb81m7u4i6ekuvzqHcNrEPs4bEMqhHhEK2iIiINBsFbmk5tVWQ/fHJkF240zneIRr6TnECdp8pEBF7wS9VUePi3QwnZH+QVURtvZueUaF8++LezB4ay5CekQrZIiIi0iIUuKX5uN3OPOy978G+NSfbRPyDIX4MTH3MCdndB1/QduoNqmpdvJdZyLIt+azJKqTG5SYmIoRvjk1g1tBYRvSKUsgWERGRFqfALU2rLN8J13vXOG8ri5zj3QbCqO85K9nx4yCoaeZXV9fV835WIUu35vNeRiHH6+rpGh7MDaPimTU0lpHxnfDzU8gWERER71HglgvzdW0iYV2d9pC+lzjj+5qgTaRBdV09H+4qYvm2fN7ZeZjK2nq6hAVx9ciezB7ag4sSO+OvkC0iIiKthAK3nBu3Gw5vOxmws9dDfa3TJpIwFoZd74TsboOapE2kQa3LzUd7ili2JZ+3dx6mvMZFpw6BzB3ek9lDYxnduzMB/k33eiIiIiJNRYFbzuxEm8h7TtCuOuIc7zYIRt3mBOz4sU3WJtKgrt7Nx3uLWbYlj7d2FFBW7SIiJIDLhsQwe2gPxvbtQqBCtoiIiLRyCtzyZbVVcPDjkzOxizKc42FdIenSk20iF7iF+ul8uq+YexZu4VDJccKDA5g2qDtzhvbg4qRoggIUskVERMR3KHBLozYRT8DO/qRRm8g4GH6jc7FjE7eJfBVXvZun3t3N02v2kNAljL9+cySTk7sSHODfrK8rIiIi0lwUuNursrwvbjrT0CbSfTCMvt254DFh3AVvOnMuco5WseCVzWw4eIxrRsbx2NxBhAXrn6iIiIj4NqWZ9qK20tMmsuaUNpFuLdYmcjpLt+Tx08XbwMJTN4xg7rAeXqlDREREpKkpcLdVbjcUbD15sWNDm0hAiHOB4/AbPZvODAIvbgZTWePi0SU7eHVDLiPio3jq+hH06ty0F1+KiIiIeJMCd1vyhTaRNVBV7BxvaBNpmCbSgm0ip7P9UCl3vbSJ/cWV/HBKEndP7aepIyIiItLmKHD7shNtIg3TRDKd42HdIGlaozaR7l4t81Rut+W5dfv5zapMuoQF85/vjmFs3y7eLktERESkWShw+5KGNpGGgJ3z6ck2kYRxMOImz6YzA73aJnI6heXV3PvqVj7cVcT0gd35zdVD6RQW5O2yRERERJqNAndr56pxwvWON2DP243aRIbA6DsatYmEeLXMs7Emq5D7Xt1CebWLJ+YN5sZR8ZhW+h8DERERkaaiwN0auWqdHuwdiyFzOdSUQUgU9J/pTBTpMxk6dvN2lWetxlXPb1Zm8dy6/aTEhPOf742hf/dwb5clIiIi0iIUuFuL+jpnHvaOxZC5DKpLISQSBsyFQfOgzyTwD/R2ledsT2EFd720iZ35Zdw8NoEHLx9ASKA2sREREZH2Q4Hbm+rrYP+HTsjOWArVJRAcCSmzPCF7MgT4Zn+ztZaF6Tk8umQnIYF+PPutNKYObF0Xb4qIiIi0BAXullbvggNrT4bs40chKPxkyO47BQKCvV3lBSmtquOni7exfFs+Fyd14ffXDad7ROvvMRcRERFpDgrcLaHeBQfXeUL2EufCx6COkHy5J2Rf4hMXPZ6Nzw8cZcHLmzlcVs39M1O4fWIf/Px0YaSIiIi0XwrczcVd78zIbgjZlUUQGAbJlzkhO+nSVrMBTVNw1bt5es0ennp3N706d+C1O8cxvFeUt8sSERER8ToF7qbkrne2UN+xGHa+CZWFENjBmS4yaB4kTYWgtrdt+aGS4yx4eROfHzjGVSN68viVg+kYrH9aIiIiIqDAfeHcbmcDmoaQXVEAAaHQf7oTsvtNh6Awb1fZbFZsy+eB17fitvDk/OFcOaKnt0sSERERaVUUuM+H2w2H0mH7Iidkl+c5uz32m+YJ2TMguKO3q2xWVbUuHl+6k5c/z2FYryieun44CV3a7n8sRERERM6XAvfZshYObXBWsne8AWW54B/sCdm/gP4zILh9bOay/VApd728if1HKvn+5L78eFp/Av39vF2WiIiISKukwH061kLexpMhuzQH/IOcXuypjzi92SER3q6yxVhreW7dAX6zMpNOYYH8+zujGZcU7e2yRERERFo1Be5TWQv5mz0hezGUZINfoDO675KfO1NGQiK9XWWLO1JRw72vbuH9rCKmDujO/1wzlM5hvrkpj4iIiEhLUuAGJ2QXbD0Zso8dAL8A6DMFJj0AKZdDaCdvV+k1H+4q4p6FWyirruMXVwzipjEJGKPZ2iIiIiJno/0Gbmvh8PaTIfvoPjD+znbqE+51dn7s0NnbVXpVrcvNb9/K5G9r99O/e0f+9d1RpMS0nxYaERERkabQvgK3tVC482TILt7jhOzeE+HiBZAyG8K6eLvKVmFfUQV3vbyJ7YfKuGlMPD+fNZCQQH9vlyUiIiLic9pH4C7MPBmyj2SB8YPECTD2hzBgDoTpwr8G1lpe3ZDLo0t2EBTgxzPfHMn0QTHeLktERETEZ7XdwF2062TILsoADCSOh9G3wYAroGNXb1fY6pQer+Nni7exbGs+Y/t04Q/zhxMTGeLtskRERER8WtsK3Ef2nAzZhTsAAwnj4PLfwYC5EN7d2xW2WhsOHuWulzZTUFbNfTOSuWNSX/z9dGGkiIiIyIXy/cBdvPfknOzD25xj8WPhsv9xQnZErFfLa+3q3Zb/W7OH//fubnpEhfDqHWNJjW+/E1lEREREmprvBu6Kw/CXCc44P4Beo2Hmr52QHdnTu7X5iLyS4yx4ZTOf7T/KFcN78IsrBxMREujtskRERETaFN8N3GV5EJAMM34FA6+AyDhvV+RTVm3P5/7Xt+Gqd/O/1w7jqtSemq0tIiIi0gx8N3B3HwTffcfbVfic47X1PL5sJy99ls3QuEieun4EidFh3i5LREREpM3yO9MdjDHPGWMKjTHbGx37rTEm0xiz1Riz2BgT5TnexRizxhhTYYx5+pTnGWmM2WaM2WOMecp4llONMcHGmFc8xz81xiSeVeX+2lb8XO3MK2PO0x/x0mfZ3D6pD6/dMU5hW0RERKSZnTFwA88DM0859jYw2Fo7FNgFPOg5Xg08BNz7Fc/zZ+A2oJ/nT8Nzfgc4Zq1NAv4A/OYc6pezYK3l+XX7ufJP6yg9Xse/vjOaBy8bQFDA2Xz6RURERORCnDFxWWs/BI6ecmy1tdbl+fATIM5zvNJa+xFO8D7BGBMLRFhr11trLfAicKXn5iuAFzzvvwZcatRM3GSKK2r4zgvpPLp0J+OToll19wTG99NGPyIiIiItpSl6uG8FXjnDfXoCuY0+zvUca7gtB8Ba6zLGlAJdgCNNUFu7tnZ3Efcs3ELp8ToenTOQm8cl6sJIERERkRZ2QYHbGPMzwAX8+0x3/Ypj9ixuO/X1bsNpSyE+Pv4sq2x/al1u/nd1Fn/9cB9J3Try4q2jGBAb4e2yRERERNql8w7cxpibgdnApZ42kdPJxdN24hEH5DW6rReQa4wJACI5pYWlgbX2GeAZgLS0tDO9Zrt04Egld728ia25pdw4Op6HZg0kNMjf22WJiIiItFvnFbiNMTOB+4FJ1tqqM93fWptvjCk3xowBPgW+BfzRc/MS4GZgPXAN8N5ZBHg5hbWWRRsP8fCb2wnw9+MvN6Uyc7B22RQRERHxtjMGbmPMS8BkINoYkws8gjOVJBh429MT/Im19g7P/Q8AEUCQMeZKYLq1didwJ87Ek1BgpecPwN+Bfxpj9uCsbF/fNKfWfpRV1/HQG9t5c3Meo3p35sn5w+kRFertskREREQEML66mJyWlmbT09O9XYbXbcw+xt0vbyKvpJoFl/bj+1OS8PfThZEiIiIiTcUYs8Fam3a+j/fdnSbbuXq35S8f7OX3b+8iJiKEhbePYWRCZ2+XJSIiIiKnUOD2QQWl1Sx4ZROf7DvK7KGxPDFvCJGhgd4uS0RERES+ggK3j1m9o4D/en0rtS43v71mKNeMjNNsbREREZFWTIHbR1TX1fPL5Tv51yfZDO4ZwVPXj6BP147eLktEREREzkCBuxWz1rI1t5SlW/JYujWPw2U1fG9Cb+6bkUJQgJ+3yxMRERGRs6DA3QrtOlzuhOwteRworiLQ3zCpf1f+cF1vxiVFe7s8ERERETkHCtytRM7RKpZ4QnZmQTl+Bsb27cKdk/syc1AskR10UaSIiIiIL1Lg9qLCsmqWbc1nyZY8NueUAJAaH8WjcwZy+dBYuoWHeLdAEREREblgCtwtrKSqlpXbC1i6JY9P9hXjtjAgNoL7Z6Ywe2gsvTp38HaJIiIiItKEFLhbQGWNi3cyDrNkcx4f7i6irt7SOzqMH17Sj7nDYknqFu7tEkVERESkmShwN5MaVz3vZxWxZEse72YcprrOTWxkCLeMS2TusJ4M7hmh+dkiIiIi7YACdxNy1bv5eG8xS7bk8daOAsqrXXQOC+KakXHMHdaTtIRO+PkpZIuIiIi0JwrcF8jttmzIPsbSLXms2JbPkYpawoMDmD4ohrnDezCubxcC/TUzW0RERKS9UuA+D9ZaduSVnZiVnVdaTXCAH1MHdGfOsFgmJ3cjJNDf22WKiIiISCugwH0O9hZVsGSzs+vjvqJKAvwME/t35b6ZyUwbGEPHYP11ioiIiMgXKSGewaGS4ydWsnfklWEMjO7dme+O78Nlg2PoFBbk7RJFREREpBVT4P4KReU1rNyez5LNeaQfPAbAsF5RPDR7ILOHxtI9QhvSiIiIiMjZUeD2KD1ex1s7nA1p1u05gttCcvdw7puRzOyhsSR0CfN2iSIiIiLig9p14D5eW+9sSLMljw+yiqitdxPfuQN3Tu7L3GE9SY7RhjQiIiIicmHaXeCudbn5cFcRS7fm8fbOw1TV1tMtPJibxiQwd3gPhsVFakMaEREREWky7SJw17stn+5zNqRZub2A0uN1RHUI5IrhPZk7rAejenfGXxvSiIiIiEgzaLOB21rLppwSlmzOY/m2fIrKawgL8nc2pBnWg4uTogkK0IY0IiIiItK82lTgttaSWVDOEs8Yv9xjxwkK8OOS5G7MGdaDS1K6ERqkDWlEREREpOW0icB94EglS7fksWRLHrsLK/D3M4xPimbB1P5MH9SdiJBAb5coIiIiIu2UzwbuunrLs2v3sWRLHltzSwEYldiZX1w5mMsHx9ClY7CXKxQRERER8eHAnVlQxi+XZzCkZyQ/u3wAs4bG0iMq1NtliYiIiIh8gc8G7u7hIbz3k0n06drR26WIiIiIiHwtnx3T0S0iWGFbRERERFo9nw3cIiIiIiK+QIFbRERERKQZKXCLiIiIiDQjBW4RERERkWakwC0iIiIi0owUuEVEREREmpECt4iIiIhIM1LgFhERERFpRgrcIiIiIiLNSIFbRERERKQZGWutt2s4L8aYciDL23W0sEig1NtFtDCdc/vQ3s65vZ0v6JzbC51z+9Aez7mftTbyfB8c0JSVtLAsa22at4toScaYZ6y1t3m7jpakc24f2ts5t7fzBZ1ze6Fzbh/a6zlfyOPVUuJblnq7AC/QObcP7e2c29v5gs65vdA5tw8653Pkyy0l6e1thVtEREREfI8vr3Bf0NK+iIiIiEhL8NkVbhERERERX+DLK9xtjjGm3hizudGfxNPc931jjE+31BhjrDHmn40+DjDGFBljlnmzrpZgjJnnOf8Ub9fSnNrz5xjAGFPh7Rq85Uzn3ka+h7WLr+PGjDE/M8bsMMZs9fycGu3tmlqCMSbOGPOmMWa3MWavMeb/GWOCTnP/BcaYDi1ZY1Px/Jv+30Yf32uMedSLJTW7RvlrhzFmizHmHmNMk2ZkBe7W5bi1dnijPwe8XVAzqwQGG2NCPR9PAw6dyxMYY3x10s4NwEfA9efyIGOMf/OU02wu+HMs0oqd19exrzLGjAVmA6nW2qHAVCDHu1U1P2OMARYBb1hr+wH9gY7AE6d52ALAJwM3UANcZYyJ9nYhLaghfw3C+Tl1OfBIU75Aqw/c7XmFCMAYM9IY84ExZoMx5i1jTGyjm28yxnxsjNlujBnltSIvzEpgluf9G4CXGm4wxozynN8mz9tkz/FbjDGvGmOWAqtbvuQLY4zpCFwMfAfPD2pjzGRjzIfGmMXGmJ3GmL80/O/aGFNhjHncGPMpMNZ7lZ+38/kcrzXGDG90v3XGmKEtWXRT8XxulzX6+GljzC2e9w8YYx4zxmw0xmxrayulpzt3X3ear+Ov+1xfbozJNMZ8ZIx5ykd/yxMLHLHW1gBYa49Ya/O+7ueU57cYT7aBn1OXANXW2n8AWGvrgR8Dtxpjwowxv/N8/W41xvzIGHMX0ANYY4xZ48W6z5cL5zq5H596gzEmwRjzrudc3zXGxBtjIj3fyxp+ZnUwxuQYYwJbuvCmYK0tBG4Dfmgc/saY3xpjPvec9+0N9zXG/Jfnc7/FGPPr0z1vqw/c7UyoOdlOstjzj/WPwDXW2pHAc3zxf9Rh1tpxwPc9t/mil4HrjTEhwFDg00a3ZQITrbUjgIeBXzW6bSxws7X2khartOlcCayy1u4CjhpjUj3HRwE/AYYAfYGrPMfDgO3W2tHW2o9autgmcD6f42eBWwCMMf2BYGvt1haruGUdsdamAn8G7vV2MXLWruSrv46/xPNv/6/AZdba8UDXlimxya0Gehljdhlj/mSMmdROfk4NAjY0PmCtLQOyge8CvYERnlX/f1trnwLygCnW2iktXWwT+T/gG8aYUzd6eRp4seFcgaestaXAFmCS5z5zgLestXUtVm0Ts9buw8nI3XD+U11qrb0IuAj4njGmtzHmMpzvA6OttcOA/zndc/pE4DbGdPT8T6phFegKz/FEY0yGMeZvxum7WW1O/uraFzVuKZkHJAODgbeNMZuBnwNxje7/EoC19kMgwhgT1cL1XjBPiErEWflcccrNkcCrxpjtwB9wvuk1eNtae7RFimx6N+CEUDxvb/C8/5m1dp9n9eQlYLzneD3wesuW2HTO83P8KjDb88P8VuD5FinWOxZ53m7A+XsS3/B1X8dfJQXYZ63d7/n4pdPct9Wy1lYAI3FW/4qAV4DbaeM/pwADfNWECQNMBP5irXUB+PDPpS/w/IfiReCuU24aC/zH8/4/Oflz6hVgvuf96z0f+zrjeTsd+Jbn3/enQBegH05L1T+stVVw5s+9r/S/VgPzrLVlxukp+sQYs8RzWz/gBmvt94wxC4GrgX95q9AmZoAd1tqvayM49RuAr46cWQL8DpiM8w+5wS+ANdbaeca5gPT9RrdVtlRxTckY0wXn15ODjTEW8Mf5vK3g6z+f1Z4Q7svO6XNsra0yxrwNXAFcB/jyxXUuvri4EXLK7TWet/X4zvfks3Wmc/dJp/k6XsJXn6+hjfB8L3ofeN8Ysw34AW3/59QOnGxxgjEmAugF7MM3z+lsPAlsBP5xmvs0nPsS4L+NMZ1x/lP2XvOW1ryMMX1wvicX4nz9/sha+9Yp95nJOXzufWKFG+dkf2WM2Qq8A/QEuntu22+t3ex5v62tEGUBXY1zoQrGmEBjTONV3vme4+Nxft1R6oUam8JzwOPW2m2nHI/k5AV2t7RoRc3nGpxfxyVYaxOttb2A/TirBKM8v6byw/nc+mL7yNc5n8/xs8BTwOc+vmp0EBhojAn2/Hr2Um8X1ILa6rl/3dcxfPX5ZgJ9zMnJU/PxQcaYZGNMv0aHhgMZtP2fU+8CHYwx34ITF6//L85v3lYDdxjPBfyewAlQDoS3fKlNx/N9dyFOS0WDjzl5kfA38Pyc8vz24zPg/wHLfHmRyBjTFfgL8LR1Zme/BdzZ0JNujOlvjAnD+dzfajzTaBp97r+Sr6ymfAOn522ktbbOGHOAkysHNY3uVw/4ckvJF1hra40x1wBPeb55B+D8j3OH5y7HjDEfAxE4v3b3SdbaXJwv0lP9D/CCMeYefPx/y43cAJx6YcXrwJ3Aes9tQ4APgcUtW1rzOZ/PsbV2gzGmjNOvrrRanh/ANdbaHM9v37YCu4FN3q2s+bWDc/+6r+MbcQLKF87XWnvcGPN9YJUx5ghOMPFFHYE/etpCXMAenPaSZ2jDP6estdYYMw/4kzHmIZzFyhXAT3FyR39gqzGmDvgbTp/zM8BKY0y+D/dxg/Mfix82+vgu4DljzH04bUXfbnTbKzjtgJNbrLqmE+ppGQnE+bf9T+D3ntuexVnM3WiMMTjnfaW1dpVxLu5PN8bUcvLfxFdq9RvfGGdKyc+AJGvtj4wxU3B+MPf23GWZtXaw5773Ah2ttY96pViR82SMmQzca62d7eVSWg1jTA+cX12nWGvdXi7nnBljhgF/s9b66mSG89aez/3rGGM6WmsrPD+w/w/Yba39g7frak7GmPdxvq+le7sWEW9r1S0lDaskOFfCphlj0nFWuzO9WpiINCvPr24/BX7mo2H7DpyLxX7u7VpaWns+9zP4nmcFbQdOK9VfvVuOiLSkVr3CrVUSEREREfF1rXaFW6skIiIiItIWtOoVbhERERERX9dqV7hFRERERNqCVhO4jTG9jDFrPDtH7jDG3O053tkY87YxZrfnbSfP8S6e+1cYY54+5bneN8ZkmZPbpHfzxjmJiIiIiLSawI0z9/An1toBwBjgB8aYgcADwLvW2n44w+cf8Ny/GngIuPdrnu8bjbZJL2zm2kVEREREvlKrCdzW2nxr7UbP++U4u1f1xNna+QXP3V4ArvTcp9Ja+xFO8BYRERERaZVaTeBuzLP97QicObzdrbX54IRy4GzbQ/7haSd5yLPRgIiIiIhIi2t1gdsY0xFni9wF1tqy83yab1hrhwATPH++2VT1iYiIiIici1YVuI0xgThh+9/W2kWew4eNMbGe22OBM/ZjW2sPed6WA/8BtHGOiIiIiHhFqwncnraPvwMZ1trfN7ppCXCz5/2bgTfP8DwBxphoz/uBwGxge9NXLCIiIiJyZq1m4xtjzHhgLbANcHsO/xSnj3shEA9kA9daa496HnMAiACCgBJgOnAQ+BAIBPyBd4B7rLX1LXQqIiIiIiIntJrALSIiIiLSFrWalhIRERERkbZIgVtEREREpBkpcIuIiIiINCMFbhERERGRZqTALSIiIiLSjBS4RUR8nDGm3hiz2RizwxizxRhzjzHmtN/fjTGJxpgbW6pGEZH2TIFbRMT3HbfWDrfWDgKmAZcDj5zhMYmAAreISAvQHG4RER9njKmw1nZs9HEf4HMgGkgA/gmEeW7+obX2Y2PMJ8AAYD/wAvAU8GtgMhAM/J+19q8tdhIiIm2YAreIiI87NXB7jh0DUoBywG2trTbG9ANestamGWMmA/daa2d77n8b0M1a+0tjTDCwDmdn3/0teS4iIm1RgLcLEBGRZmE8bwOBp40xw4F6oP/X3H86MNQYc43n40igH84KuIiIXAAFbhGRNsbTUlIPFOL0ch8GhuFct1P9dQ8DfmStfatFihQRaUd00aSISBtijOkK/AV42jo9g5FAvrXWDXwT8PfctRwIb/TQt4A7jTGBnufpb4wJQ0RELphWuEVEfF+oMWYzTvuIC+ciyd97bvsT8Lox5lpgDVDpOb4VcBljtgDPA/8PZ3LJRmOMAYqAK1umfBGRtk0XTYqIiIiINCO1lIiIiIiINCMFbhERERGRZqTALSIiIiLSjBS4RURERESakQK3iIiIiEgzUuAWEREREWlGCtwiIiIiIs3o/wMV+G8twy8Y6QAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 864x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "df['Spending'][-n_obs:].plot(figsize=(12,5),legend=True).autoscale(axis='x',tight=True)\n",
    "df_forecast['SpendingForecast'].plot(legend=True);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "id": "c17b705b-5652-4471-96ee-fd935bee3afb",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "M2 VAR(5) RMSE: 43.710\n"
     ]
    }
   ],
   "source": [
    "RMSE1 = rmse(df['Money'][-n_obs:], df_forecast['MoneyForecast'])\n",
    "print(f'M2 VAR(5) RMSE: {RMSE1:.3f}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "id": "55d68c71-f1e5-48c6-ac3e-f21026a6de7f",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Spending VAR(5) RMSE: 37.001\n"
     ]
    }
   ],
   "source": [
    "RMSE2 = rmse(df['Spending'][-n_obs:], df_forecast['SpendingForecast'])\n",
    "print(f'Spending VAR(5) RMSE: {RMSE2:.3f}')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fd783884-b199-485c-a6a7-989892cbbe13",
   "metadata": {},
   "source": [
    "Nota-se que o modelo mais simples performou muito melhor com os dados."
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}