10. K-means no Dolar Futuro.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 10. K-means no Dólar Futuro\n",
    " \n",
    "\n",
    "#### Etapas\n",
    "\n",
    "- Passo 0 - Assitir às aulas anteriores de K-means\n",
    "- Passo 1 - Carregar as bibliotecas\n",
    "- Passo 2 - Carregar a base de dados\n",
    "- Passo 3 - Tratamento dos dados\n",
    "- Passo 4 - Construçao dos alvos e variaveis\n",
    "- Passo 5 - Preparaçao da base para treinamento e teste\n",
    "- Passo 6 - Avaliaçao dos numeros de clusters necessàrios\n",
    "- Passo 7 - Avaliaçao dos Resultados\n",
    "- Passo 8 - Analise bivariada\n",
    "- Passo 9 - Retreinamento do K-means\n",
    "- Passo 10 - Re-avaliaçao dos Resultados\n",
    "\n",
    " \n",
    "\n",
    "________________________________________________________\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Passo 1 - Carregar as bibliotecas"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Carregando as bibliotecas\n",
    "\n",
    "import time\n",
    "import matplotlib.pyplot as plt\n",
    "from sklearn import datasets\n",
    "from sklearn.decomposition import PCA\n",
    "from sklearn.model_selection import train_test_split, GridSearchCV, StratifiedKFold\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "from sklearn.metrics import classification_report, confusion_matrix, precision_score, accuracy_score, precision_recall_curve, auc, make_scorer, recall_score\n",
    "from sklearn.cluster import KMeans\n",
    "import statsmodels.api as sm\n",
    "import pandas as pd\n",
    "import numpy as np\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Para remover os warnings\n",
    "import sys\n",
    "\n",
    "if not sys.warnoptions:\n",
    "    import warnings\n",
    "    warnings.simplefilter(\"ignore\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "______________________________________________________"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Passo 2 - Carregar a base de dados"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "3854"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Carregando a base de dados e fazendo os checks iniciais\n",
    "\n",
    "# Muda o diretorio de trabalho\n",
    "import os\n",
    "os.chdir(\"C:\\\\Users\\\\engle\\\\OM Na Pratica\\\\Finanças Quantitativas\\\\Bases de Dados\")\n",
    "\n",
    "df1 = pd.read_csv(\"WDOFUT.csv\", sep = \";\")\n",
    "\n",
    "df1.shape[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "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>Data</th>\n",
       "      <th>Abertura</th>\n",
       "      <th>Maxima</th>\n",
       "      <th>Minima</th>\n",
       "      <th>Fechamento</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>17/05/2005</td>\n",
       "      <td>2487.0</td>\n",
       "      <td>2502.0</td>\n",
       "      <td>2487.0</td>\n",
       "      <td>2494.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>18/05/2005</td>\n",
       "      <td>2484.0</td>\n",
       "      <td>2484.0</td>\n",
       "      <td>2469.0</td>\n",
       "      <td>2470.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>19/05/2005</td>\n",
       "      <td>2455.0</td>\n",
       "      <td>2458.0</td>\n",
       "      <td>2450.0</td>\n",
       "      <td>2455.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>20/05/2005</td>\n",
       "      <td>2460.0</td>\n",
       "      <td>2463.0</td>\n",
       "      <td>2451.0</td>\n",
       "      <td>2451.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>23/05/2005</td>\n",
       "      <td>2443.0</td>\n",
       "      <td>2443.0</td>\n",
       "      <td>2432.0</td>\n",
       "      <td>2433.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         Data  Abertura  Maxima  Minima  Fechamento\n",
       "0  17/05/2005    2487.0  2502.0  2487.0      2494.0\n",
       "1  18/05/2005    2484.0  2484.0  2469.0      2470.0\n",
       "2  19/05/2005    2455.0  2458.0  2450.0      2455.0\n",
       "3  20/05/2005    2460.0  2463.0  2451.0      2451.0\n",
       "4  23/05/2005    2443.0  2443.0  2432.0      2433.0"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df1.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "______________________________________________________"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Passo 3 - Tratamento dos dados"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Ajustando a data - precisamos transforma-la nos indices do data frame\n",
    "\n",
    "df1[\"Data\"] = pd.to_datetime(df1[\"Data\"]).dt.normalize()\n",
    "\n",
    "df1 = df1.set_index(\"Data\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Abertura</th>\n",
       "      <th>Maxima</th>\n",
       "      <th>Minima</th>\n",
       "      <th>Fechamento</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Data</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2005-05-17</th>\n",
       "      <td>2487.0</td>\n",
       "      <td>2502.0</td>\n",
       "      <td>2487.0</td>\n",
       "      <td>2494.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2005-05-18</th>\n",
       "      <td>2484.0</td>\n",
       "      <td>2484.0</td>\n",
       "      <td>2469.0</td>\n",
       "      <td>2470.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2005-05-19</th>\n",
       "      <td>2455.0</td>\n",
       "      <td>2458.0</td>\n",
       "      <td>2450.0</td>\n",
       "      <td>2455.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2005-05-20</th>\n",
       "      <td>2460.0</td>\n",
       "      <td>2463.0</td>\n",
       "      <td>2451.0</td>\n",
       "      <td>2451.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2005-05-23</th>\n",
       "      <td>2443.0</td>\n",
       "      <td>2443.0</td>\n",
       "      <td>2432.0</td>\n",
       "      <td>2433.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            Abertura  Maxima  Minima  Fechamento\n",
       "Data                                            \n",
       "2005-05-17    2487.0  2502.0  2487.0      2494.0\n",
       "2005-05-18    2484.0  2484.0  2469.0      2470.0\n",
       "2005-05-19    2455.0  2458.0  2450.0      2455.0\n",
       "2005-05-20    2460.0  2463.0  2451.0      2451.0\n",
       "2005-05-23    2443.0  2443.0  2432.0      2433.0"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df1.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Abertura</th>\n",
       "      <th>Maxima</th>\n",
       "      <th>Minima</th>\n",
       "      <th>Fechamento</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>3854.000000</td>\n",
       "      <td>3854.000000</td>\n",
       "      <td>3854.000000</td>\n",
       "      <td>3854.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>2695.424453</td>\n",
       "      <td>2714.421069</td>\n",
       "      <td>2677.683342</td>\n",
       "      <td>2695.850519</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>996.551114</td>\n",
       "      <td>1006.839596</td>\n",
       "      <td>986.645656</td>\n",
       "      <td>997.148535</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>1534.500000</td>\n",
       "      <td>1542.000000</td>\n",
       "      <td>1530.000000</td>\n",
       "      <td>1541.500000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>1910.000000</td>\n",
       "      <td>1921.375000</td>\n",
       "      <td>1896.250000</td>\n",
       "      <td>1909.125000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>2275.500000</td>\n",
       "      <td>2292.750000</td>\n",
       "      <td>2260.000000</td>\n",
       "      <td>2274.750000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>3312.125000</td>\n",
       "      <td>3333.000000</td>\n",
       "      <td>3292.750000</td>\n",
       "      <td>3313.750000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>5927.500000</td>\n",
       "      <td>5977.500000</td>\n",
       "      <td>5823.500000</td>\n",
       "      <td>5894.500000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          Abertura       Maxima       Minima   Fechamento\n",
       "count  3854.000000  3854.000000  3854.000000  3854.000000\n",
       "mean   2695.424453  2714.421069  2677.683342  2695.850519\n",
       "std     996.551114  1006.839596   986.645656   997.148535\n",
       "min    1534.500000  1542.000000  1530.000000  1541.500000\n",
       "25%    1910.000000  1921.375000  1896.250000  1909.125000\n",
       "50%    2275.500000  2292.750000  2260.000000  2274.750000\n",
       "75%    3312.125000  3333.000000  3292.750000  3313.750000\n",
       "max    5927.500000  5977.500000  5823.500000  5894.500000"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df1.describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "______________________________________________________"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Passo 4 - Construçao dos alvos e variaveis"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Construcao dos alvos - Vamos trabalhar com um horizonte mais longo, de 5 periodos?\n",
    "\n",
    "periodos = 1\n",
    "\n",
    "# lag de X periodos\n",
    "df1[\"Lag1\"] = df1[\"Fechamento\"].shift(-periodos)\n",
    "\n",
    "# Alvo 1 - Retorno\n",
    "df1[\"Retorno\"] = df1[\"Fechamento\"].pct_change(periodos)\n",
    "df1[\"Alvo1\"] = df1[\"Retorno\"].shift(-periodos)\n",
    "\n",
    "# Variaçao em Pontos do alvo\n",
    "df1[\"Pontos\"] = (df1[\"Fechamento\"] - df1[\"Fechamento\"].shift(1))\n",
    "df1[\"Pontos\"] = df1[\"Pontos\"].shift(-periodos)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count    3853.000000\n",
       "mean        0.000246\n",
       "std         0.011053\n",
       "min        -0.079897\n",
       "25%        -0.005851\n",
       "50%         0.000000\n",
       "75%         0.005831\n",
       "max         0.078145\n",
       "Name: Retorno, dtype: float64"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df1[\"Retorno\"].describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1296x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize = (18,5))\n",
    "plt.hist(df1[\"Retorno\"], bins = 50\n",
    "        , alpha = 0.45\n",
    "        , histtype = \"stepfilled\"\n",
    "        , color = \"darkgreen\"\n",
    "        , edgecolor = \"none\"\n",
    "        , label = \"Retornos\");\n",
    "\n",
    "plt.legend()\n",
    "plt.title(\"Distribuiçao Retornos\");"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Criacao do alvo\n",
    "df1[\"Alvo_Bin\"] = np.where(df1['Alvo1'] > df1[\"Retorno\"].describe()[6]/2 , 1\n",
    "                           , np.where(df1['Alvo1'] < df1[\"Retorno\"].describe()[4]/2, -1, 0))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1296x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize = (18,5))\n",
    "plt.hist(df1[\"Alvo_Bin\"], bins = 5\n",
    "        , alpha = 0.45\n",
    "        , histtype = \"stepfilled\"\n",
    "        , color = \"darkgreen\"\n",
    "        , edgecolor = \"none\"\n",
    "        , label = \"Retornos\");\n",
    "\n",
    "plt.legend()\n",
    "plt.title(\"Distribuiçao Retornos\");"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "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>Abertura</th>\n",
       "      <th>Maxima</th>\n",
       "      <th>Minima</th>\n",
       "      <th>Fechamento</th>\n",
       "      <th>Lag1</th>\n",
       "      <th>Retorno</th>\n",
       "      <th>Alvo1</th>\n",
       "      <th>Pontos</th>\n",
       "      <th>Alvo_Bin</th>\n",
       "      <th>std5</th>\n",
       "      <th>...</th>\n",
       "      <th>a</th>\n",
       "      <th>m</th>\n",
       "      <th>f</th>\n",
       "      <th>T</th>\n",
       "      <th>cat</th>\n",
       "      <th>w</th>\n",
       "      <th>k</th>\n",
       "      <th>tau</th>\n",
       "      <th>M</th>\n",
       "      <th>g</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Data</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2005-06-28</th>\n",
       "      <td>2369.0</td>\n",
       "      <td>2375.0</td>\n",
       "      <td>2369.0</td>\n",
       "      <td>2374.0</td>\n",
       "      <td>2360.0</td>\n",
       "      <td>-0.002940</td>\n",
       "      <td>-0.005897</td>\n",
       "      <td>-14.0</td>\n",
       "      <td>-1</td>\n",
       "      <td>0.007640</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000274</td>\n",
       "      <td>-0.004085</td>\n",
       "      <td>-1.119103e-06</td>\n",
       "      <td>-3.144358e-09</td>\n",
       "      <td>-0.000229</td>\n",
       "      <td>-1.119103e-06</td>\n",
       "      <td>0.000005</td>\n",
       "      <td>2.562563e-10</td>\n",
       "      <td>-0.000005</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2005-06-29</th>\n",
       "      <td>2365.0</td>\n",
       "      <td>2365.0</td>\n",
       "      <td>2350.0</td>\n",
       "      <td>2360.0</td>\n",
       "      <td>2350.5</td>\n",
       "      <td>-0.005897</td>\n",
       "      <td>-0.004025</td>\n",
       "      <td>-9.5</td>\n",
       "      <td>-1</td>\n",
       "      <td>0.007206</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.000768</td>\n",
       "      <td>-0.014596</td>\n",
       "      <td>1.121397e-05</td>\n",
       "      <td>-3.225396e-08</td>\n",
       "      <td>-0.001740</td>\n",
       "      <td>1.121395e-05</td>\n",
       "      <td>0.000127</td>\n",
       "      <td>-1.950994e-08</td>\n",
       "      <td>0.000031</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2005-06-30</th>\n",
       "      <td>2346.0</td>\n",
       "      <td>2356.0</td>\n",
       "      <td>2346.0</td>\n",
       "      <td>2350.5</td>\n",
       "      <td>2371.0</td>\n",
       "      <td>-0.004025</td>\n",
       "      <td>0.008722</td>\n",
       "      <td>20.5</td>\n",
       "      <td>1</td>\n",
       "      <td>0.003145</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.001223</td>\n",
       "      <td>-0.027390</td>\n",
       "      <td>3.350148e-05</td>\n",
       "      <td>-8.966265e-08</td>\n",
       "      <td>-0.004673</td>\n",
       "      <td>3.350111e-05</td>\n",
       "      <td>0.000640</td>\n",
       "      <td>-1.565469e-07</td>\n",
       "      <td>0.000070</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2005-01-07</th>\n",
       "      <td>2357.0</td>\n",
       "      <td>2382.0</td>\n",
       "      <td>2357.0</td>\n",
       "      <td>2371.0</td>\n",
       "      <td>2392.0</td>\n",
       "      <td>0.008722</td>\n",
       "      <td>0.008857</td>\n",
       "      <td>21.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.005860</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000912</td>\n",
       "      <td>-0.007907</td>\n",
       "      <td>-7.211111e-06</td>\n",
       "      <td>-6.002671e-08</td>\n",
       "      <td>-0.003326</td>\n",
       "      <td>-7.211071e-06</td>\n",
       "      <td>0.000131</td>\n",
       "      <td>2.398154e-08</td>\n",
       "      <td>-0.000031</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2005-04-07</th>\n",
       "      <td>2400.0</td>\n",
       "      <td>2403.0</td>\n",
       "      <td>2392.0</td>\n",
       "      <td>2392.0</td>\n",
       "      <td>2404.0</td>\n",
       "      <td>0.008857</td>\n",
       "      <td>0.005017</td>\n",
       "      <td>12.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.007240</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000756</td>\n",
       "      <td>0.004716</td>\n",
       "      <td>3.563098e-06</td>\n",
       "      <td>1.502823e-08</td>\n",
       "      <td>-0.000828</td>\n",
       "      <td>3.563097e-06</td>\n",
       "      <td>-0.000020</td>\n",
       "      <td>-2.950982e-09</td>\n",
       "      <td>0.000012</td>\n",
       "      <td>0.000022</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2005-05-07</th>\n",
       "      <td>2400.0</td>\n",
       "      <td>2404.0</td>\n",
       "      <td>2384.0</td>\n",
       "      <td>2404.0</td>\n",
       "      <td>2415.0</td>\n",
       "      <td>0.005017</td>\n",
       "      <td>0.004576</td>\n",
       "      <td>11.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.007045</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000070</td>\n",
       "      <td>0.012673</td>\n",
       "      <td>8.885036e-07</td>\n",
       "      <td>1.604573e-08</td>\n",
       "      <td>0.001531</td>\n",
       "      <td>8.885026e-07</td>\n",
       "      <td>0.000097</td>\n",
       "      <td>1.360462e-09</td>\n",
       "      <td>0.000020</td>\n",
       "      <td>0.000161</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2005-06-07</th>\n",
       "      <td>2406.0</td>\n",
       "      <td>2422.0</td>\n",
       "      <td>2406.0</td>\n",
       "      <td>2415.0</td>\n",
       "      <td>2396.0</td>\n",
       "      <td>0.004576</td>\n",
       "      <td>-0.007867</td>\n",
       "      <td>-19.0</td>\n",
       "      <td>-1</td>\n",
       "      <td>0.005236</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000839</td>\n",
       "      <td>0.023146</td>\n",
       "      <td>1.942769e-05</td>\n",
       "      <td>5.077316e-08</td>\n",
       "      <td>0.003714</td>\n",
       "      <td>1.942756e-05</td>\n",
       "      <td>0.000430</td>\n",
       "      <td>7.215345e-08</td>\n",
       "      <td>0.000048</td>\n",
       "      <td>0.000536</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2005-07-07</th>\n",
       "      <td>2426.0</td>\n",
       "      <td>2427.0</td>\n",
       "      <td>2396.0</td>\n",
       "      <td>2396.0</td>\n",
       "      <td>2388.0</td>\n",
       "      <td>-0.007867</td>\n",
       "      <td>-0.003339</td>\n",
       "      <td>-8.0</td>\n",
       "      <td>-1</td>\n",
       "      <td>0.006855</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000358</td>\n",
       "      <td>0.019304</td>\n",
       "      <td>6.911906e-06</td>\n",
       "      <td>5.698978e-09</td>\n",
       "      <td>0.005434</td>\n",
       "      <td>6.911804e-06</td>\n",
       "      <td>0.000524</td>\n",
       "      <td>3.756012e-08</td>\n",
       "      <td>-0.000015</td>\n",
       "      <td>0.000373</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2005-08-07</th>\n",
       "      <td>2406.0</td>\n",
       "      <td>2406.0</td>\n",
       "      <td>2379.0</td>\n",
       "      <td>2388.0</td>\n",
       "      <td>2359.0</td>\n",
       "      <td>-0.003339</td>\n",
       "      <td>-0.012144</td>\n",
       "      <td>-29.0</td>\n",
       "      <td>-1</td>\n",
       "      <td>0.006839</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.001262</td>\n",
       "      <td>0.007243</td>\n",
       "      <td>-9.138882e-06</td>\n",
       "      <td>2.107069e-08</td>\n",
       "      <td>0.002116</td>\n",
       "      <td>-9.138861e-06</td>\n",
       "      <td>0.000077</td>\n",
       "      <td>-1.934141e-08</td>\n",
       "      <td>-0.000017</td>\n",
       "      <td>0.000052</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2005-11-07</th>\n",
       "      <td>2388.0</td>\n",
       "      <td>2388.0</td>\n",
       "      <td>2359.0</td>\n",
       "      <td>2359.0</td>\n",
       "      <td>2355.0</td>\n",
       "      <td>-0.012144</td>\n",
       "      <td>-0.001696</td>\n",
       "      <td>-4.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.007563</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.001345</td>\n",
       "      <td>-0.013758</td>\n",
       "      <td>1.850387e-05</td>\n",
       "      <td>-1.213581e-07</td>\n",
       "      <td>-0.000323</td>\n",
       "      <td>1.850387e-05</td>\n",
       "      <td>0.000022</td>\n",
       "      <td>-5.972912e-09</td>\n",
       "      <td>0.000058</td>\n",
       "      <td>0.000189</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>10 rows × 34 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "            Abertura  Maxima  Minima  Fechamento    Lag1   Retorno     Alvo1  \\\n",
       "Data                                                                           \n",
       "2005-06-28    2369.0  2375.0  2369.0      2374.0  2360.0 -0.002940 -0.005897   \n",
       "2005-06-29    2365.0  2365.0  2350.0      2360.0  2350.5 -0.005897 -0.004025   \n",
       "2005-06-30    2346.0  2356.0  2346.0      2350.5  2371.0 -0.004025  0.008722   \n",
       "2005-01-07    2357.0  2382.0  2357.0      2371.0  2392.0  0.008722  0.008857   \n",
       "2005-04-07    2400.0  2403.0  2392.0      2392.0  2404.0  0.008857  0.005017   \n",
       "2005-05-07    2400.0  2404.0  2384.0      2404.0  2415.0  0.005017  0.004576   \n",
       "2005-06-07    2406.0  2422.0  2406.0      2415.0  2396.0  0.004576 -0.007867   \n",
       "2005-07-07    2426.0  2427.0  2396.0      2396.0  2388.0 -0.007867 -0.003339   \n",
       "2005-08-07    2406.0  2406.0  2379.0      2388.0  2359.0 -0.003339 -0.012144   \n",
       "2005-11-07    2388.0  2388.0  2359.0      2359.0  2355.0 -0.012144 -0.001696   \n",
       "\n",
       "            Pontos  Alvo_Bin      std5  ...         a         m             f  \\\n",
       "Data                                    ...                                     \n",
       "2005-06-28   -14.0        -1  0.007640  ...  0.000274 -0.004085 -1.119103e-06   \n",
       "2005-06-29    -9.5        -1  0.007206  ... -0.000768 -0.014596  1.121397e-05   \n",
       "2005-06-30    20.5         1  0.003145  ... -0.001223 -0.027390  3.350148e-05   \n",
       "2005-01-07    21.0         1  0.005860  ...  0.000912 -0.007907 -7.211111e-06   \n",
       "2005-04-07    12.0         1  0.007240  ...  0.000756  0.004716  3.563098e-06   \n",
       "2005-05-07    11.0         1  0.007045  ...  0.000070  0.012673  8.885036e-07   \n",
       "2005-06-07   -19.0        -1  0.005236  ...  0.000839  0.023146  1.942769e-05   \n",
       "2005-07-07    -8.0        -1  0.006855  ...  0.000358  0.019304  6.911906e-06   \n",
       "2005-08-07   -29.0        -1  0.006839  ... -0.001262  0.007243 -9.138882e-06   \n",
       "2005-11-07    -4.0         0  0.007563  ... -0.001345 -0.013758  1.850387e-05   \n",
       "\n",
       "                       T       cat             w         k           tau  \\\n",
       "Data                                                                       \n",
       "2005-06-28 -3.144358e-09 -0.000229 -1.119103e-06  0.000005  2.562563e-10   \n",
       "2005-06-29 -3.225396e-08 -0.001740  1.121395e-05  0.000127 -1.950994e-08   \n",
       "2005-06-30 -8.966265e-08 -0.004673  3.350111e-05  0.000640 -1.565469e-07   \n",
       "2005-01-07 -6.002671e-08 -0.003326 -7.211071e-06  0.000131  2.398154e-08   \n",
       "2005-04-07  1.502823e-08 -0.000828  3.563097e-06 -0.000020 -2.950982e-09   \n",
       "2005-05-07  1.604573e-08  0.001531  8.885026e-07  0.000097  1.360462e-09   \n",
       "2005-06-07  5.077316e-08  0.003714  1.942756e-05  0.000430  7.215345e-08   \n",
       "2005-07-07  5.698978e-09  0.005434  6.911804e-06  0.000524  3.756012e-08   \n",
       "2005-08-07  2.107069e-08  0.002116 -9.138861e-06  0.000077 -1.934141e-08   \n",
       "2005-11-07 -1.213581e-07 -0.000323  1.850387e-05  0.000022 -5.972912e-09   \n",
       "\n",
       "                   M         g  \n",
       "Data                            \n",
       "2005-06-28 -0.000005       NaN  \n",
       "2005-06-29  0.000031       NaN  \n",
       "2005-06-30  0.000070       NaN  \n",
       "2005-01-07 -0.000031       NaN  \n",
       "2005-04-07  0.000012  0.000022  \n",
       "2005-05-07  0.000020  0.000161  \n",
       "2005-06-07  0.000048  0.000536  \n",
       "2005-07-07 -0.000015  0.000373  \n",
       "2005-08-07 -0.000017  0.000052  \n",
       "2005-11-07  0.000058  0.000189  \n",
       "\n",
       "[10 rows x 34 columns]"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Construçao das Variaveis\n",
    "\n",
    "# Desvio Padrao de 5 dias\n",
    "df1[\"std5\"] = df1[\"Retorno\"].rolling(5).std()\n",
    "\n",
    "# Desvio Padrao de 10 dias\n",
    "df1[\"std10\"] = df1[\"Retorno\"].rolling(10).std()\n",
    "\n",
    "# Desvio Padrao de 15 dias\n",
    "df1[\"std15\"] = df1[\"Retorno\"].rolling(15).std()\n",
    "\n",
    "# Proporçao do corpo do candle em relacao ao range do dia\n",
    "df1[\"prop\"] = (df1[\"Fechamento\"]-df1[\"Abertura\"])/(df1[\"Maxima\"]-df1[\"Minima\"])\n",
    "\n",
    "# Direçao do dia atual\n",
    "df1[\"dir_D\"] = np.where(df1['Fechamento'] > df1['Abertura'] , 1, 0)\n",
    "\n",
    "# Direçao D-1\n",
    "df1[\"dir_D-1\"] = df1[\"dir_D\"].shift(1)\n",
    "\n",
    "# Direçao D-2\n",
    "df1[\"dir_D-2\"] = df1[\"dir_D\"].shift(2)\n",
    "\n",
    "# Direçao D-3\n",
    "df1[\"dir_D-3\"] = df1[\"dir_D\"].shift(3)\n",
    "\n",
    "# Media Movel de 15 dias std5\n",
    "df1[\"mm_std5\"] = df1[\"std5\"].rolling(15).mean()\n",
    "\n",
    "# Media Movel de 15 dias std5\n",
    "df1[\"mm_std10\"] = df1[\"std10\"].rolling(15).mean()\n",
    "\n",
    "# Media Movel de 15 dias std5\n",
    "df1[\"mm_std15\"] = df1[\"std15\"].rolling(15).mean()\n",
    "\n",
    "# RSL std5\n",
    "df1[\"RSL_std5\"] = (df1[\"std5\"]/df1[\"std5\"].rolling(15).mean())-1\n",
    "\n",
    "# RSL std10\n",
    "df1[\"RSL_std10\"] = (df1[\"std10\"]/df1[\"std10\"].rolling(15).mean())-1\n",
    "\n",
    "# RSL std15\n",
    "df1[\"RSL_std15\"] = (df1[\"std15\"]/df1[\"std15\"].rolling(15).mean())-1\n",
    "\n",
    "# Aplicando a fisica\n",
    "\n",
    "p = 5 \n",
    "\n",
    "# Velocidade de p dias\n",
    "df1[\"v\"] = (df1[\"Retorno\"] - df1[\"Retorno\"].shift(p))/p\n",
    "\n",
    "# Aceleraçao de p dias\n",
    "df1[\"a\"] = (df1[\"v\"] - df1[\"v\"].shift(p))/p\n",
    "\n",
    "# Força\n",
    "# Calculando a massa\n",
    "df1[\"m\"] = df1[\"Retorno\"].rolling(p).sum()\n",
    "df1[\"f\"] = df1[\"m\"]*df1[\"a\"]\n",
    "\n",
    "# Energia cinética\n",
    "df1[\"T\"] = 0.5*df1[\"m\"]*df1[\"v\"]*df1[\"v\"]\n",
    "\n",
    "\n",
    "# Trabalho \n",
    "# cateto_oposto\n",
    "cat_op = df1[\"Retorno\"].rolling(p).sum()-df1[\"Retorno\"].rolling(1).sum()\n",
    "cat_ad = p\n",
    "df1[\"cat\"] = cat_op/cat_ad\n",
    "df1 = df1.dropna(axis = 0)\n",
    "df1[\"w\"] = df1[\"f\"]*np.cos(np.arctan(df1[\"cat\"]))\n",
    "\n",
    "# Energia potencial\n",
    "df1[\"k\"] = cat_op*df1[\"m\"]\n",
    "\n",
    "# Torque\n",
    "df1[\"tau\"] = df1[\"f\"]*np.sin(np.arctan(df1[\"cat\"]))\n",
    "\n",
    "# Momentum\n",
    "df1[\"M\"] = df1[\"m\"]*df1[\"v\"]\n",
    "\n",
    "# Gravidade\n",
    "df1[\"g\"] = df1[\"m\"]*df1[\"Retorno\"].rolling(p).sum()/(df1[\"m\"]/df1[\"Retorno\"].rolling(p).sum())*(df1[\"m\"]/df1[\"Retorno\"].rolling(p).sum())\n",
    "\n",
    "df1.head(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Filtrando os valores missing\n",
    "\n",
    "df1_filtrado = df1.dropna(axis = 0) \n",
    "\n",
    "df1_filtrado= df1_filtrado.drop([\"std5\",\"std10\",\"std15\",\"mm_std5\",\"mm_std10\",\"mm_std15\"], axis = 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "corte = 10\n",
    "\n",
    "df1_filtrado[\"RSL_std5\"] = pd.qcut(df1_filtrado[\"RSL_std5\"], corte, labels = False)\n",
    "df1_filtrado[\"RSL_std10\"] = pd.qcut(df1_filtrado[\"RSL_std10\"], corte, labels = False)\n",
    "df1_filtrado[\"RSL_std15\"] = pd.qcut(df1_filtrado[\"RSL_std15\"], corte, labels = False)\n",
    "df1_filtrado[\"v\"] = pd.qcut(df1_filtrado[\"v\"], corte, labels = False)\n",
    "df1_filtrado[\"a\"] = pd.qcut(df1_filtrado[\"a\"], corte, labels = False)\n",
    "df1_filtrado[\"m\"] = pd.qcut(df1_filtrado[\"m\"], corte, labels = False)\n",
    "df1_filtrado[\"f\"] = pd.qcut(df1_filtrado[\"f\"], corte, labels = False)\n",
    "df1_filtrado[\"T\"] = pd.qcut(df1_filtrado[\"T\"], corte, labels = False)\n",
    "df1_filtrado[\"w\"] = pd.qcut(df1_filtrado[\"w\"], corte, labels = False)\n",
    "df1_filtrado[\"k\"] = pd.qcut(df1_filtrado[\"k\"], corte, labels = False)\n",
    "df1_filtrado[\"tau\"] = pd.qcut(df1_filtrado[\"tau\"], corte, labels = False)\n",
    "df1_filtrado[\"M\"] = pd.qcut(df1_filtrado[\"M\"], corte, labels = False)\n",
    "df1_filtrado[\"g\"] = pd.qcut(df1_filtrado[\"g\"], corte, labels = False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>Alvo_Bin</th>\n",
       "      <th>-1</th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>RSL_std5</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>3.294979</td>\n",
       "      <td>2.955021</td>\n",
       "      <td>3.765690</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>3.817992</td>\n",
       "      <td>2.771967</td>\n",
       "      <td>3.399582</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>4.184100</td>\n",
       "      <td>2.588912</td>\n",
       "      <td>3.216527</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3.713389</td>\n",
       "      <td>2.641213</td>\n",
       "      <td>3.661088</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>3.817992</td>\n",
       "      <td>2.693515</td>\n",
       "      <td>3.478033</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>3.687238</td>\n",
       "      <td>2.536611</td>\n",
       "      <td>3.765690</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>3.347280</td>\n",
       "      <td>2.981172</td>\n",
       "      <td>3.687238</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>3.713389</td>\n",
       "      <td>2.981172</td>\n",
       "      <td>3.294979</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>3.530335</td>\n",
       "      <td>2.641213</td>\n",
       "      <td>3.817992</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>3.687238</td>\n",
       "      <td>2.353556</td>\n",
       "      <td>3.974895</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Alvo_Bin        -1         0         1\n",
       "RSL_std5                              \n",
       "0         3.294979  2.955021  3.765690\n",
       "1         3.817992  2.771967  3.399582\n",
       "2         4.184100  2.588912  3.216527\n",
       "3         3.713389  2.641213  3.661088\n",
       "4         3.817992  2.693515  3.478033\n",
       "5         3.687238  2.536611  3.765690\n",
       "6         3.347280  2.981172  3.687238\n",
       "7         3.713389  2.981172  3.294979\n",
       "8         3.530335  2.641213  3.817992\n",
       "9         3.687238  2.353556  3.974895"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.crosstab(pd.qcut(df1[\"RSL_std5\"], corte, labels = False), df1[\"Alvo_Bin\"])/df1.shape[0]*100"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>Alvo_Bin</th>\n",
       "      <th>-1</th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>RSL_std5</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>7.112971</td>\n",
       "      <td>5.726987</td>\n",
       "      <td>7.165272</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>7.897490</td>\n",
       "      <td>5.230126</td>\n",
       "      <td>6.877615</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>7.505230</td>\n",
       "      <td>5.230126</td>\n",
       "      <td>7.243724</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>7.060669</td>\n",
       "      <td>5.962343</td>\n",
       "      <td>6.982218</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>7.217573</td>\n",
       "      <td>4.994770</td>\n",
       "      <td>7.792887</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Alvo_Bin        -1         0         1\n",
       "RSL_std5                              \n",
       "0         7.112971  5.726987  7.165272\n",
       "1         7.897490  5.230126  6.877615\n",
       "2         7.505230  5.230126  7.243724\n",
       "3         7.060669  5.962343  6.982218\n",
       "4         7.217573  4.994770  7.792887"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.crosstab(pd.qcut(df1[\"RSL_std5\"], 5, labels = False), df1[\"Alvo_Bin\"])/df1.shape[0]*100"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(pd.crosstab(pd.qcut(df1[\"RSL_std10\"], 20, labels = False), df1[\"Alvo_Bin\"])/df1.shape[0]*100);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "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>Abertura</th>\n",
       "      <th>Maxima</th>\n",
       "      <th>Minima</th>\n",
       "      <th>Fechamento</th>\n",
       "      <th>Lag1</th>\n",
       "      <th>Retorno</th>\n",
       "      <th>Alvo1</th>\n",
       "      <th>Pontos</th>\n",
       "      <th>Alvo_Bin</th>\n",
       "      <th>prop</th>\n",
       "      <th>...</th>\n",
       "      <th>a</th>\n",
       "      <th>m</th>\n",
       "      <th>f</th>\n",
       "      <th>T</th>\n",
       "      <th>cat</th>\n",
       "      <th>w</th>\n",
       "      <th>k</th>\n",
       "      <th>tau</th>\n",
       "      <th>M</th>\n",
       "      <th>g</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Data</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2020-08-12</th>\n",
       "      <td>5121.5</td>\n",
       "      <td>5139.5</td>\n",
       "      <td>5063.0</td>\n",
       "      <td>5120.0</td>\n",
       "      <td>5172.5</td>\n",
       "      <td>0.004020</td>\n",
       "      <td>0.010254</td>\n",
       "      <td>52.5</td>\n",
       "      <td>1</td>\n",
       "      <td>-0.019608</td>\n",
       "      <td>...</td>\n",
       "      <td>9</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>-0.004200</td>\n",
       "      <td>0</td>\n",
       "      <td>7</td>\n",
       "      <td>9</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-09-12</th>\n",
       "      <td>5100.5</td>\n",
       "      <td>5198.0</td>\n",
       "      <td>5087.0</td>\n",
       "      <td>5172.5</td>\n",
       "      <td>5027.5</td>\n",
       "      <td>0.010254</td>\n",
       "      <td>-0.028033</td>\n",
       "      <td>-145.0</td>\n",
       "      <td>-1</td>\n",
       "      <td>0.648649</td>\n",
       "      <td>...</td>\n",
       "      <td>4</td>\n",
       "      <td>3</td>\n",
       "      <td>5</td>\n",
       "      <td>2</td>\n",
       "      <td>-0.003837</td>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-10-12</th>\n",
       "      <td>5112.0</td>\n",
       "      <td>5133.0</td>\n",
       "      <td>5016.0</td>\n",
       "      <td>5027.5</td>\n",
       "      <td>5070.5</td>\n",
       "      <td>-0.028033</td>\n",
       "      <td>0.008553</td>\n",
       "      <td>43.0</td>\n",
       "      <td>1</td>\n",
       "      <td>-0.722222</td>\n",
       "      <td>...</td>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000762</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>8</td>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-11-12</th>\n",
       "      <td>5055.0</td>\n",
       "      <td>5089.0</td>\n",
       "      <td>5029.5</td>\n",
       "      <td>5070.5</td>\n",
       "      <td>5117.5</td>\n",
       "      <td>0.008553</td>\n",
       "      <td>0.009269</td>\n",
       "      <td>47.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.260504</td>\n",
       "      <td>...</td>\n",
       "      <td>6</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>-0.005058</td>\n",
       "      <td>1</td>\n",
       "      <td>7</td>\n",
       "      <td>8</td>\n",
       "      <td>1</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-12-14</th>\n",
       "      <td>5039.0</td>\n",
       "      <td>5131.0</td>\n",
       "      <td>5010.5</td>\n",
       "      <td>5117.5</td>\n",
       "      <td>5089.5</td>\n",
       "      <td>0.009269</td>\n",
       "      <td>-0.005471</td>\n",
       "      <td>-28.0</td>\n",
       "      <td>-1</td>\n",
       "      <td>0.651452</td>\n",
       "      <td>...</td>\n",
       "      <td>8</td>\n",
       "      <td>5</td>\n",
       "      <td>6</td>\n",
       "      <td>8</td>\n",
       "      <td>-0.001041</td>\n",
       "      <td>6</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>6</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 28 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "            Abertura  Maxima  Minima  Fechamento    Lag1   Retorno     Alvo1  \\\n",
       "Data                                                                           \n",
       "2020-08-12    5121.5  5139.5  5063.0      5120.0  5172.5  0.004020  0.010254   \n",
       "2020-09-12    5100.5  5198.0  5087.0      5172.5  5027.5  0.010254 -0.028033   \n",
       "2020-10-12    5112.0  5133.0  5016.0      5027.5  5070.5 -0.028033  0.008553   \n",
       "2020-11-12    5055.0  5089.0  5029.5      5070.5  5117.5  0.008553  0.009269   \n",
       "2020-12-14    5039.0  5131.0  5010.5      5117.5  5089.5  0.009269 -0.005471   \n",
       "\n",
       "            Pontos  Alvo_Bin      prop  ...  a  m  f  T       cat  w  k  tau  \\\n",
       "Data                                    ...                                    \n",
       "2020-08-12    52.5         1 -0.019608  ...  9  1  0  0 -0.004200  0  7    9   \n",
       "2020-09-12  -145.0        -1  0.648649  ...  4  3  5  2 -0.003837  5  5    2   \n",
       "2020-10-12    43.0         1 -0.722222  ...  5  1  3  0  0.000762  3  0    4   \n",
       "2020-11-12    47.0         1  0.260504  ...  6  1  1  2 -0.005058  1  7    8   \n",
       "2020-12-14   -28.0        -1  0.651452  ...  8  5  6  8 -0.001041  6  0    3   \n",
       "\n",
       "            M  g  \n",
       "Data              \n",
       "2020-08-12  0  6  \n",
       "2020-09-12  1  3  \n",
       "2020-10-12  8  7  \n",
       "2020-11-12  1  6  \n",
       "2020-12-14  6  1  \n",
       "\n",
       "[5 rows x 28 columns]"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df1_filtrado.tail()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['Abertura', 'Maxima', 'Minima', 'Fechamento', 'Lag1', 'Retorno',\n",
       "       'Alvo1', 'Pontos', 'Alvo_Bin', 'prop', 'dir_D', 'dir_D-1', 'dir_D-2',\n",
       "       'dir_D-3', 'RSL_std5', 'RSL_std10', 'RSL_std15', 'v', 'a', 'm', 'f',\n",
       "       'T', 'cat', 'w', 'k', 'tau', 'M', 'g'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df1_filtrado.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "df1_filtrado = df1_filtrado.drop([\"m\",\"cat\",\"M\"], axis=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "______________________________________________________"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Passo 5 - Preparaçao da base para treinamento e teste"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Separando os dados entre treinamento e teste\n",
    "\n",
    "# Vamos treinar o modelo de 2005 à 2012\n",
    "start_train = \"2005-08-07\"\n",
    "end_train = \"2012-12-31\"\n",
    "\n",
    "# Vamos testar o modelo de 2013 à 2020\n",
    "start_test = \"2013-01-01\"\n",
    "end_test = \"2020-12-08\"\n",
    "\n",
    "df1_train1 = df1_filtrado.loc[start_train : end_train]\n",
    "\n",
    "df1_test1 = df1_filtrado.loc[start_test : end_test]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "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>Abertura</th>\n",
       "      <th>Maxima</th>\n",
       "      <th>Minima</th>\n",
       "      <th>Fechamento</th>\n",
       "      <th>Lag1</th>\n",
       "      <th>Retorno</th>\n",
       "      <th>Alvo1</th>\n",
       "      <th>Pontos</th>\n",
       "      <th>Alvo_Bin</th>\n",
       "      <th>prop</th>\n",
       "      <th>...</th>\n",
       "      <th>RSL_std10</th>\n",
       "      <th>RSL_std15</th>\n",
       "      <th>v</th>\n",
       "      <th>a</th>\n",
       "      <th>f</th>\n",
       "      <th>T</th>\n",
       "      <th>w</th>\n",
       "      <th>k</th>\n",
       "      <th>tau</th>\n",
       "      <th>g</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Data</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2013-02-01</th>\n",
       "      <td>2041.0</td>\n",
       "      <td>2056.5</td>\n",
       "      <td>2041.0</td>\n",
       "      <td>2055.5</td>\n",
       "      <td>2057.0</td>\n",
       "      <td>-0.001700</td>\n",
       "      <td>0.000730</td>\n",
       "      <td>1.5</td>\n",
       "      <td>0</td>\n",
       "      <td>0.935484</td>\n",
       "      <td>...</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>6</td>\n",
       "      <td>2</td>\n",
       "      <td>4</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>6</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2013-03-01</th>\n",
       "      <td>2057.0</td>\n",
       "      <td>2059.5</td>\n",
       "      <td>2043.5</td>\n",
       "      <td>2057.0</td>\n",
       "      <td>2043.5</td>\n",
       "      <td>0.000730</td>\n",
       "      <td>-0.006563</td>\n",
       "      <td>-13.5</td>\n",
       "      <td>-1</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>6</td>\n",
       "      <td>3</td>\n",
       "      <td>6</td>\n",
       "      <td>5</td>\n",
       "      <td>2</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2013-04-01</th>\n",
       "      <td>2057.5</td>\n",
       "      <td>2059.0</td>\n",
       "      <td>2041.0</td>\n",
       "      <td>2043.5</td>\n",
       "      <td>2039.0</td>\n",
       "      <td>-0.006563</td>\n",
       "      <td>-0.002202</td>\n",
       "      <td>-4.5</td>\n",
       "      <td>0</td>\n",
       "      <td>-0.777778</td>\n",
       "      <td>...</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>7</td>\n",
       "      <td>8</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2013-07-01</th>\n",
       "      <td>2044.0</td>\n",
       "      <td>2045.5</td>\n",
       "      <td>2033.0</td>\n",
       "      <td>2039.0</td>\n",
       "      <td>2050.0</td>\n",
       "      <td>-0.002202</td>\n",
       "      <td>0.005395</td>\n",
       "      <td>11.0</td>\n",
       "      <td>1</td>\n",
       "      <td>-0.400000</td>\n",
       "      <td>...</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>5</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2013-08-01</th>\n",
       "      <td>2036.5</td>\n",
       "      <td>2050.0</td>\n",
       "      <td>2029.5</td>\n",
       "      <td>2050.0</td>\n",
       "      <td>2050.0</td>\n",
       "      <td>0.005395</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.658537</td>\n",
       "      <td>...</td>\n",
       "      <td>4</td>\n",
       "      <td>2</td>\n",
       "      <td>4</td>\n",
       "      <td>2</td>\n",
       "      <td>6</td>\n",
       "      <td>4</td>\n",
       "      <td>6</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 25 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "            Abertura  Maxima  Minima  Fechamento    Lag1   Retorno     Alvo1  \\\n",
       "Data                                                                           \n",
       "2013-02-01    2041.0  2056.5  2041.0      2055.5  2057.0 -0.001700  0.000730   \n",
       "2013-03-01    2057.0  2059.5  2043.5      2057.0  2043.5  0.000730 -0.006563   \n",
       "2013-04-01    2057.5  2059.0  2041.0      2043.5  2039.0 -0.006563 -0.002202   \n",
       "2013-07-01    2044.0  2045.5  2033.0      2039.0  2050.0 -0.002202  0.005395   \n",
       "2013-08-01    2036.5  2050.0  2029.5      2050.0  2050.0  0.005395  0.000000   \n",
       "\n",
       "            Pontos  Alvo_Bin      prop  ...  RSL_std10  RSL_std15  v  a  f  T  \\\n",
       "Data                                    ...                                     \n",
       "2013-02-01     1.5         0  0.935484  ...          3          0  4  6  2  4   \n",
       "2013-03-01   -13.5        -1  0.000000  ...          4          0  3  3  6  3   \n",
       "2013-04-01    -4.5         0 -0.777778  ...          3          1  7  8  2  3   \n",
       "2013-07-01    11.0         1 -0.400000  ...          3          1  5  4  4  4   \n",
       "2013-08-01     0.0         0  0.658537  ...          4          2  4  2  6  4   \n",
       "\n",
       "            w  k  tau  g  \n",
       "Data                      \n",
       "2013-02-01  2  3    6  3  \n",
       "2013-03-01  6  5    2  4  \n",
       "2013-04-01  2  0    3  1  \n",
       "2013-07-01  4  1    4  1  \n",
       "2013-08-01  6  3    2  2  \n",
       "\n",
       "[5 rows x 25 columns]"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df1_test1.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['Abertura', 'Maxima', 'Minima', 'Fechamento', 'Lag1', 'Retorno',\n",
       "       'Alvo1', 'Pontos', 'Alvo_Bin', 'prop', 'dir_D', 'dir_D-1', 'dir_D-2',\n",
       "       'dir_D-3', 'RSL_std5', 'RSL_std10', 'RSL_std15', 'v', 'a', 'f', 'T',\n",
       "       'w', 'k', 'tau', 'g'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df1_filtrado.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Separando os dados com as variaveis em x e o alvo em y\n",
    " \n",
    "x_train1 = df1_train1.iloc[:, 9:24]\n",
    "y_train1 = df1_train1['Alvo_Bin'].values.reshape(-1, 1)\n",
    "\n",
    "x_test1 = df1_test1.iloc[:, 9:24]\n",
    "y_test1 = df1_test1['Alvo_Bin'].values.reshape(-1, 1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "______________________________________________________"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Passo 6 - Avaliaçao dos numeros de clusters necessàrios\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "min_clusters = 2\n",
    "max_clusters = 20\n",
    "inertias = np.zeros(shape = (max_clusters - min_clusters + 1,))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "for i in range(min_clusters, max_clusters + 1):\n",
    "    km = KMeans(n_clusters = i, random_state = 42)\n",
    "    km.fit(x_train1)\n",
    "    inertias[i - min_clusters] = km.inertia_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize = (10, 6))\n",
    "plt.plot(range(min_clusters, max_clusters + 1), inertias, color = \"b\")\n",
    "plt.title(\"Avaliaçao do numero de clusters\")\n",
    "plt.xlabel(\"# de clusters\")\n",
    "plt.ylabel(\"Soma da distancia quadrada\")\n",
    "plt.annotate(\"# de clusters escolhido\", xy = (6, 68000), xytext = (7, 90000),\n",
    "            arrowprops = dict(facecolor = \"red\", shrink = 0.05),\n",
    "            );"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [],
   "source": [
    "kmodel = KMeans(n_clusters = 6, random_state = 42)\n",
    "clusters = kmodel.fit(x_train1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([5, 0, 0, ..., 0, 4, 4])"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Acessando os clusters\n",
    "\n",
    "clusters.labels_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Aplicando os clusters na base de treinamento e na base de teste\n",
    "# Atribuido-os à base de treinamento e teste\n",
    "\n",
    "train_clusters = pd.DataFrame(clusters.labels_,columns = [\"Cluster\"])\n",
    "train_clusters[\"Label\"] = y_train1\n",
    "train_clusters[\"Alvo1\"] = df1_train1['Alvo1'].values.reshape(-1, 1)\n",
    "\n",
    "\n",
    "fitted_clusters = kmodel.predict(x_test1)\n",
    "test_clusters = pd.DataFrame(fitted_clusters,columns = [\"Cluster\"])\n",
    "test_clusters[\"Label\"] = y_test1\n",
    "test_clusters[\"Alvo1\"] = df1_test1['Alvo1'].values.reshape(-1, 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "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>Cluster</th>\n",
       "      <th>Label</th>\n",
       "      <th>Alvo1</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>5</td>\n",
       "      <td>-1</td>\n",
       "      <td>-0.012144</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>-0.001696</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.006369</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>-1</td>\n",
       "      <td>-0.019108</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "      <td>-1</td>\n",
       "      <td>-0.003030</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Cluster  Label     Alvo1\n",
       "0        5     -1 -0.012144\n",
       "1        0      0 -0.001696\n",
       "2        0      1  0.006369\n",
       "3        4     -1 -0.019108\n",
       "4        0     -1 -0.003030"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_clusters.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "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>Cluster</th>\n",
       "      <th>Label</th>\n",
       "      <th>Alvo1</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000730</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "      <td>-1</td>\n",
       "      <td>-0.006563</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>-0.002202</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>0.005395</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Cluster  Label     Alvo1\n",
       "0        2      0  0.000730\n",
       "1        0     -1 -0.006563\n",
       "2        2      0 -0.002202\n",
       "3        2      1  0.005395\n",
       "4        2      0  0.000000"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_clusters.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "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>Label</th>\n",
       "      <th>-1</th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Cluster</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>102</td>\n",
       "      <td>97</td>\n",
       "      <td>104</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>107</td>\n",
       "      <td>60</td>\n",
       "      <td>95</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>181</td>\n",
       "      <td>190</td>\n",
       "      <td>128</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>102</td>\n",
       "      <td>108</td>\n",
       "      <td>107</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>93</td>\n",
       "      <td>63</td>\n",
       "      <td>85</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>86</td>\n",
       "      <td>26</td>\n",
       "      <td>77</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Label     -1    0    1\n",
       "Cluster               \n",
       "0        102   97  104\n",
       "1        107   60   95\n",
       "2        181  190  128\n",
       "3        102  108  107\n",
       "4         93   63   85\n",
       "5         86   26   77"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.crosstab(train_clusters[\"Cluster\"], train_clusters[\"Label\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Cluster\n",
       "0    0.000535\n",
       "1   -0.000230\n",
       "2   -0.000586\n",
       "3    0.000693\n",
       "4   -0.000246\n",
       "5    0.000283\n",
       "Name: Alvo1, dtype: float64"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_clusters.groupby(\"Cluster\")[\"Alvo1\"].mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Cluster\n",
       "0    0.162087\n",
       "1   -0.060139\n",
       "2   -0.292625\n",
       "3    0.219776\n",
       "4   -0.059273\n",
       "5    0.053431\n",
       "Name: Alvo1, dtype: float64"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_clusters.groupby(\"Cluster\")[\"Alvo1\"].sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "______________________________________________________"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Passo 7 - Avaliaçao dos Resultados"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Calculando a performance - para toda a base\n",
    "\n",
    "y_pred_all = kmodel.predict(df1_filtrado.iloc[:, 9:24])\n",
    "df1_filtrado.loc[: , \"Clusters\"] = y_pred_all"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [],
   "source": [
    "df1_filtrado.loc[: , \"Previsto\"] = np.where(((df1_filtrado.loc[: , \"Clusters\"] == 0) | (df1_filtrado.loc[: , \"Clusters\"] == 3)), 1,\n",
    "                                           np.where((df1_filtrado.loc[: , \"Clusters\"] == 2),-1\n",
    "                                                   ,0))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [],
   "source": [
    "df1_filtrado.loc[: , \"Retorno_Modelo\"] = df1_filtrado.loc[: , \"Previsto\"]*df1_filtrado[\"Alvo1\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [],
   "source": [
    "df1_filtrado.loc[: , \"Retorno_Modelo_Acumulado\"] = df1_filtrado[\"Retorno_Modelo\"].cumsum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Data\n",
       "2020-08-12    1.161336\n",
       "2020-09-12    1.189368\n",
       "2020-10-12    1.197921\n",
       "2020-11-12    1.197921\n",
       "2020-12-14    1.197921\n",
       "Name: Retorno_Modelo_Acumulado, dtype: float64"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df1_filtrado.loc[: , \"Retorno_Modelo_Acumulado\"].tail()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Faz uma copia de df1_filtrado\n",
    "df_stats = df1_filtrado.copy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Para fazer um conditional plot\n",
    "\n",
    "df1_filtrado.index.name = \"Date\"\n",
    "df1_filtrado.reset_index(inplace = True)\n",
    "\n",
    "df1_filtrado[\"Date\"] = pd.to_datetime(df1_filtrado[\"Date\"])\n",
    "\n",
    "df1_filtrado[\"train_test\"] = np.where(df1_filtrado[\"Date\"] > end_train, 1, -1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [],
   "source": [
    "base_agregada = df1_filtrado.resample(\"M\", on = \"Date\").sum()\n",
    "\n",
    "base_agregada.loc[: , \"Retorno_Modelo_Acumulado\"] = base_agregada[\"Retorno_Modelo\"].cumsum()*100"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {
    "scrolled": false
   },
   "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",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Empty DataFrame\n",
       "Columns: []\n",
       "Index: []"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x1080 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plota o grafico\n",
    "fig, ax = plt.subplots(figsize = (20,15))\n",
    "\n",
    "def grafico(group):\n",
    "    color = \"black\" if (group[\"train_test\"] < 0).all() else \"blue\"\n",
    "    lw = 4.0\n",
    "    ax.plot(group.index, group.Retorno_Modelo_Acumulado, c=color, linewidth=lw)\n",
    "    plt.xlabel(\"Ano\"\n",
    "           , fontsize = 18);\n",
    "    plt.ylabel(\"Retorno Acumulado (%)\"\n",
    "           , fontsize = 18);\n",
    "    plt.title(\"K-Means - 7 anos no DOLFUT\", fontsize = 24)\n",
    "    plt.axvline( x = end_train, color='purple', linestyle = \"--\", lw = 2)\n",
    "    setup = dict(size = 16, color = \"darkgreen\")\n",
    "    setup2 = dict(size = 14, color = \"black\")\n",
    "    plt.text(\"2009-08-01\", 90, \"Treinamento: 2005-2012\", **setup)\n",
    "    plt.text(\"2017-01-01\", 80, \"Teste: 2013-2020\", **setup)\n",
    "    plt.text(\"2010-01-01\", 86.5, \"Retorno total: +120%\", **setup2)\n",
    "    plt.text(\"2010-01-01\", 83.0, \"Retorno teste: +49%\", **setup2)\n",
    "    plt.text(\"2017-03-01\", 85.5, \"www.outspokenmarket.com\", **setup2)\n",
    "\n",
    "base_agregada.groupby((base_agregada[\"train_test\"].shift() * base_agregada[\"train_test\"] < 0).cumsum()).apply(grafico)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Counter({2: 983, 1: 648, 5: 449, 0: 616, 3: 644, 4: 480})"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import collections\n",
    "\n",
    "collections.Counter(y_pred_all)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2243"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.count_nonzero((y_pred_all == 0) | (y_pred_all == 3) | (y_pred_all == 2))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "3820"
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(y_pred_all)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [],
   "source": [
    "long = np.count_nonzero((y_pred_all == 0) | (y_pred_all == 3))\n",
    "short = np.count_nonzero(y_pred_all == 2)\n",
    "trades = np.count_nonzero((y_pred_all == 0) | (y_pred_all == 3) | (y_pred_all == 2))\n",
    "total = len(y_pred_all)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "K-Means: DolFut - 2003 à 2020\n",
      "Outspoken Market\n",
      "--------------------------\n",
      "# Pregoes:        3820\n",
      "# Trades:         2243\n",
      "--------------------------\n",
      "Tempo em trades:  58.72 %\n",
      "Tempo em Long:    32.98 %\n",
      "Tempo em Short:   25.73 %\n",
      "--------------------------\n",
      "Trades em Long:   56.17 %\n",
      "Trades em Short:  43.83 %\n"
     ]
    }
   ],
   "source": [
    "print(\"K-Means: DolFut - 2003 à 2020\")\n",
    "print(\"Outspoken Market\")\n",
    "print(\"--------------------------\")\n",
    "print(\"# Pregoes:       \", total)\n",
    "print(\"# Trades:        \", trades)\n",
    "print(\"--------------------------\")\n",
    "print(\"Tempo em trades: \", round(trades/total*100,2),\"%\")\n",
    "print(\"Tempo em Long:   \", round(long/total*100,2),\"%\")\n",
    "print(\"Tempo em Short:  \", round(short/total*100,2),\"%\")\n",
    "print(\"--------------------------\")\n",
    "print(\"Trades em Long:  \", round(long/trades*100,2),\"%\")\n",
    "print(\"Trades em Short: \", round(short/trades*100,2),\"%\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Clusters\n",
       "0    38.227070\n",
       "1     0.000000\n",
       "2    18.438902\n",
       "3    63.126171\n",
       "4     0.000000\n",
       "5     0.000000\n",
       "Name: Retorno_Modelo, dtype: float64"
      ]
     },
     "execution_count": 55,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_stats.groupby(\"Clusters\")[\"Retorno_Modelo\"].sum()*100"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Cluster\n",
       "0    0.253278\n",
       "1   -0.064995\n",
       "2    0.116478\n",
       "3    0.391563\n",
       "4    0.271128\n",
       "5    0.058095\n",
       "Name: Alvo1, dtype: float64"
      ]
     },
     "execution_count": 56,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_clusters.groupby(\"Cluster\")[\"Alvo1\"].sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "______________________________________________________"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Passo 8 - Analise bivariada"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_stats_train = df_stats.loc[start_train : end_train]\n",
    "\n",
    "df_stats_test = df_stats.loc[start_test : end_test]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['Abertura', 'Maxima', 'Minima', 'Fechamento', 'Lag1', 'Retorno',\n",
       "       'Alvo1', 'Pontos', 'Alvo_Bin', 'prop', 'dir_D', 'dir_D-1', 'dir_D-2',\n",
       "       'dir_D-3', 'RSL_std5', 'RSL_std10', 'RSL_std15', 'v', 'a', 'f', 'T',\n",
       "       'w', 'k', 'tau', 'g', 'Clusters', 'Previsto', 'Retorno_Modelo',\n",
       "       'Retorno_Modelo_Acumulado'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 58,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_stats.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "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>prop</th>\n",
       "      <th>dir_D</th>\n",
       "      <th>dir_D-1</th>\n",
       "      <th>dir_D-2</th>\n",
       "      <th>dir_D-3</th>\n",
       "      <th>RSL_std5</th>\n",
       "      <th>RSL_std10</th>\n",
       "      <th>RSL_std15</th>\n",
       "      <th>v</th>\n",
       "      <th>a</th>\n",
       "      <th>f</th>\n",
       "      <th>T</th>\n",
       "      <th>w</th>\n",
       "      <th>k</th>\n",
       "      <th>tau</th>\n",
       "      <th>g</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Clusters</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>-0.294906</td>\n",
       "      <td>0.300330</td>\n",
       "      <td>0.346535</td>\n",
       "      <td>0.372937</td>\n",
       "      <td>0.346535</td>\n",
       "      <td>3.775578</td>\n",
       "      <td>4.333333</td>\n",
       "      <td>4.376238</td>\n",
       "      <td>1.452145</td>\n",
       "      <td>1.349835</td>\n",
       "      <td>7.666667</td>\n",
       "      <td>1.039604</td>\n",
       "      <td>7.666667</td>\n",
       "      <td>5.874587</td>\n",
       "      <td>1.323432</td>\n",
       "      <td>6.270627</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>-0.007753</td>\n",
       "      <td>0.468938</td>\n",
       "      <td>0.515030</td>\n",
       "      <td>0.486974</td>\n",
       "      <td>0.472946</td>\n",
       "      <td>2.719439</td>\n",
       "      <td>1.817635</td>\n",
       "      <td>2.084168</td>\n",
       "      <td>4.663327</td>\n",
       "      <td>4.611222</td>\n",
       "      <td>4.110220</td>\n",
       "      <td>4.513026</td>\n",
       "      <td>4.110220</td>\n",
       "      <td>2.567134</td>\n",
       "      <td>4.364729</td>\n",
       "      <td>2.282565</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>-0.081234</td>\n",
       "      <td>0.425868</td>\n",
       "      <td>0.438486</td>\n",
       "      <td>0.438486</td>\n",
       "      <td>0.492114</td>\n",
       "      <td>6.946372</td>\n",
       "      <td>6.965300</td>\n",
       "      <td>6.353312</td>\n",
       "      <td>4.100946</td>\n",
       "      <td>4.186120</td>\n",
       "      <td>3.892744</td>\n",
       "      <td>4.015773</td>\n",
       "      <td>3.892744</td>\n",
       "      <td>2.239748</td>\n",
       "      <td>4.750789</td>\n",
       "      <td>2.223975</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              prop     dir_D   dir_D-1   dir_D-2   dir_D-3  RSL_std5  \\\n",
       "Clusters                                                               \n",
       "0        -0.294906  0.300330  0.346535  0.372937  0.346535  3.775578   \n",
       "2        -0.007753  0.468938  0.515030  0.486974  0.472946  2.719439   \n",
       "3        -0.081234  0.425868  0.438486  0.438486  0.492114  6.946372   \n",
       "\n",
       "          RSL_std10  RSL_std15         v         a         f         T  \\\n",
       "Clusters                                                                 \n",
       "0          4.333333   4.376238  1.452145  1.349835  7.666667  1.039604   \n",
       "2          1.817635   2.084168  4.663327  4.611222  4.110220  4.513026   \n",
       "3          6.965300   6.353312  4.100946  4.186120  3.892744  4.015773   \n",
       "\n",
       "                 w         k       tau         g  \n",
       "Clusters                                          \n",
       "0         7.666667  5.874587  1.323432  6.270627  \n",
       "2         4.110220  2.567134  4.364729  2.282565  \n",
       "3         3.892744  2.239748  4.750789  2.223975  "
      ]
     },
     "execution_count": 59,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_stats_train[df_stats_train.Clusters.isin([0,3,2])].groupby(\"Clusters\")['prop', 'dir_D', 'dir_D-1', 'dir_D-2',\n",
    "       'dir_D-3', 'RSL_std5', 'RSL_std10', 'RSL_std15', 'v', 'a', 'f', 'T',\n",
    "       'w', 'k', 'tau', 'g'].mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "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>prop</th>\n",
       "      <th>dir_D</th>\n",
       "      <th>dir_D-1</th>\n",
       "      <th>dir_D-2</th>\n",
       "      <th>dir_D-3</th>\n",
       "      <th>RSL_std5</th>\n",
       "      <th>RSL_std10</th>\n",
       "      <th>RSL_std15</th>\n",
       "      <th>v</th>\n",
       "      <th>a</th>\n",
       "      <th>f</th>\n",
       "      <th>T</th>\n",
       "      <th>w</th>\n",
       "      <th>k</th>\n",
       "      <th>tau</th>\n",
       "      <th>g</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Clusters</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>-0.317041</td>\n",
       "      <td>0.264706</td>\n",
       "      <td>0.352941</td>\n",
       "      <td>0.359477</td>\n",
       "      <td>0.326797</td>\n",
       "      <td>4.022876</td>\n",
       "      <td>4.392157</td>\n",
       "      <td>4.385621</td>\n",
       "      <td>1.444444</td>\n",
       "      <td>1.310458</td>\n",
       "      <td>7.732026</td>\n",
       "      <td>1.003268</td>\n",
       "      <td>7.732026</td>\n",
       "      <td>5.911765</td>\n",
       "      <td>1.454248</td>\n",
       "      <td>6.30719</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.019188</td>\n",
       "      <td>0.518750</td>\n",
       "      <td>0.522917</td>\n",
       "      <td>0.497917</td>\n",
       "      <td>0.550000</td>\n",
       "      <td>2.614583</td>\n",
       "      <td>1.972917</td>\n",
       "      <td>2.304167</td>\n",
       "      <td>4.658333</td>\n",
       "      <td>4.535417</td>\n",
       "      <td>4.066667</td>\n",
       "      <td>4.970833</td>\n",
       "      <td>4.066667</td>\n",
       "      <td>2.654167</td>\n",
       "      <td>4.433333</td>\n",
       "      <td>2.50625</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>-0.072823</td>\n",
       "      <td>0.450000</td>\n",
       "      <td>0.478125</td>\n",
       "      <td>0.503125</td>\n",
       "      <td>0.512500</td>\n",
       "      <td>6.825000</td>\n",
       "      <td>6.843750</td>\n",
       "      <td>6.281250</td>\n",
       "      <td>3.928125</td>\n",
       "      <td>3.950000</td>\n",
       "      <td>4.134375</td>\n",
       "      <td>3.978125</td>\n",
       "      <td>4.134375</td>\n",
       "      <td>2.259375</td>\n",
       "      <td>4.775000</td>\n",
       "      <td>2.61875</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              prop     dir_D   dir_D-1   dir_D-2   dir_D-3  RSL_std5  \\\n",
       "Clusters                                                               \n",
       "0        -0.317041  0.264706  0.352941  0.359477  0.326797  4.022876   \n",
       "2         0.019188  0.518750  0.522917  0.497917  0.550000  2.614583   \n",
       "3        -0.072823  0.450000  0.478125  0.503125  0.512500  6.825000   \n",
       "\n",
       "          RSL_std10  RSL_std15         v         a         f         T  \\\n",
       "Clusters                                                                 \n",
       "0          4.392157   4.385621  1.444444  1.310458  7.732026  1.003268   \n",
       "2          1.972917   2.304167  4.658333  4.535417  4.066667  4.970833   \n",
       "3          6.843750   6.281250  3.928125  3.950000  4.134375  3.978125   \n",
       "\n",
       "                 w         k       tau        g  \n",
       "Clusters                                         \n",
       "0         7.732026  5.911765  1.454248  6.30719  \n",
       "2         4.066667  2.654167  4.433333  2.50625  \n",
       "3         4.134375  2.259375  4.775000  2.61875  "
      ]
     },
     "execution_count": 60,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_stats_test[df_stats_test.Clusters.isin([0,3,2])].groupby(\"Clusters\")['prop', 'dir_D', 'dir_D-1', 'dir_D-2',\n",
    "       'dir_D-3', 'RSL_std5', 'RSL_std10', 'RSL_std15', 'v', 'a', 'f', 'T',\n",
    "       'w', 'k', 'tau', 'g'].mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Separando os dados com as variaveis em x e o alvo em y\n",
    " \n",
    "x_train1 = df_stats_train.iloc[:, 9:16]\n",
    "y_train1 = df_stats_train['Alvo_Bin'].values.reshape(-1, 1)\n",
    "\n",
    "x_test1 = df_stats_test.iloc[:, 9:16]\n",
    "y_test1 = df_stats_test['Alvo_Bin'].values.reshape(-1, 1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "______________________________________________________"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Passo 9 - Retreinamento do K-means\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "min_clusters = 2\n",
    "max_clusters = 20\n",
    "inertias = np.zeros(shape = (max_clusters - min_clusters + 1,))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [],
   "source": [
    "for i in range(min_clusters, max_clusters + 1):\n",
    "    km = KMeans(n_clusters = i, random_state = 42)\n",
    "    km.fit(x_train1)\n",
    "    inertias[i - min_clusters] = km.inertia_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize = (10, 6))\n",
    "plt.plot(range(min_clusters, max_clusters + 1), inertias, color = \"b\")\n",
    "plt.title(\"Avaliaçao do numero de clusters\")\n",
    "plt.xlabel(\"# de clusters\")\n",
    "plt.ylabel(\"Soma da distancia quadrada\")\n",
    "plt.annotate(\"# de clusters escolhido\", xy = (7, 6000), xytext = (7.5, 7000),\n",
    "            arrowprops = dict(facecolor = \"red\", shrink = 0.05),\n",
    "            );"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [],
   "source": [
    "kmodel = KMeans(n_clusters = 6, random_state = 42)\n",
    "clusters = kmodel.fit(x_train1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([2, 1, 5, ..., 2, 2, 1])"
      ]
     },
     "execution_count": 66,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Acessando os clusters\n",
    "\n",
    "clusters.labels_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Aplicando os clusters na base de treinamento e na base de teste\n",
    "# Atribuido-os à base de treinamento e teste\n",
    "\n",
    "train_clusters = pd.DataFrame(clusters.labels_,columns = [\"Cluster\"])\n",
    "train_clusters[\"Label\"] = y_train1\n",
    "train_clusters[\"Alvo1\"] = df1_train1['Alvo1'].values.reshape(-1, 1)\n",
    "\n",
    "\n",
    "fitted_clusters = kmodel.predict(x_test1)\n",
    "test_clusters = pd.DataFrame(fitted_clusters,columns = [\"Cluster\"])\n",
    "test_clusters[\"Label\"] = y_test1\n",
    "test_clusters[\"Alvo1\"] = df1_test1['Alvo1'].values.reshape(-1, 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "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>Label</th>\n",
       "      <th>-1</th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Cluster</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>163</td>\n",
       "      <td>129</td>\n",
       "      <td>127</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>78</td>\n",
       "      <td>58</td>\n",
       "      <td>61</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>132</td>\n",
       "      <td>121</td>\n",
       "      <td>107</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>132</td>\n",
       "      <td>76</td>\n",
       "      <td>133</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>76</td>\n",
       "      <td>74</td>\n",
       "      <td>86</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>90</td>\n",
       "      <td>86</td>\n",
       "      <td>82</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Label     -1    0    1\n",
       "Cluster               \n",
       "0        163  129  127\n",
       "1         78   58   61\n",
       "2        132  121  107\n",
       "3        132   76  133\n",
       "4         76   74   86\n",
       "5         90   86   82"
      ]
     },
     "execution_count": 68,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.crosstab(train_clusters[\"Cluster\"], train_clusters[\"Label\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Cluster\n",
       "0   -0.072615\n",
       "1   -0.103731\n",
       "2   -0.218418\n",
       "3    0.323700\n",
       "4    0.144395\n",
       "5   -0.050075\n",
       "Name: Alvo1, dtype: float64"
      ]
     },
     "execution_count": 69,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_clusters.groupby(\"Cluster\")[\"Alvo1\"].sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "______________________________________________________"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Passo 10 - Re-avaliaçao dos Resultados"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Calculando a performance - para toda a base\n",
    "\n",
    "y_pred_all = kmodel.predict(df_stats.iloc[:, 9:16])\n",
    "df_stats.loc[: , \"Clusters\"] = y_pred_all"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_stats.loc[: , \"Previsto\"] = np.where(((df_stats.loc[: , \"Clusters\"] == 3) ), 1,\n",
    "                                           np.where((df_stats.loc[: , \"Clusters\"] == 2),-1\n",
    "                                                   ,0))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_stats.loc[: , \"Retorno_Modelo\"] = df_stats.loc[: , \"Previsto\"]*df_stats[\"Alvo1\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_stats.loc[: , \"Retorno_Modelo_Acumulado\"] = df_stats[\"Retorno_Modelo\"].cumsum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Data\n",
       "2020-08-12    1.044040\n",
       "2020-09-12    1.072072\n",
       "2020-10-12    1.080625\n",
       "2020-11-12    1.089895\n",
       "2020-12-14    1.084423\n",
       "Name: Retorno_Modelo_Acumulado, dtype: float64"
      ]
     },
     "execution_count": 74,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_stats.loc[: , \"Retorno_Modelo_Acumulado\"].tail()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Para fazer um conditional plot\n",
    "\n",
    "df_stats.index.name = \"Date\"\n",
    "df_stats.reset_index(inplace = True)\n",
    "\n",
    "df_stats[\"Date\"] = pd.to_datetime(df_stats[\"Date\"])\n",
    "\n",
    "df_stats[\"train_test\"] = np.where(df_stats[\"Date\"] > end_train, 1, -1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {},
   "outputs": [],
   "source": [
    "base_agregada = df_stats.resample(\"M\", on = \"Date\").sum()\n",
    "\n",
    "base_agregada.loc[: , \"Retorno_Modelo_Acumulado\"] = base_agregada[\"Retorno_Modelo\"].cumsum()*100"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {
    "scrolled": false
   },
   "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",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Empty DataFrame\n",
       "Columns: []\n",
       "Index: []"
      ]
     },
     "execution_count": 77,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x1080 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plota o grafico\n",
    "fig, ax = plt.subplots(figsize = (20,15))\n",
    "\n",
    "def grafico(group):\n",
    "    color = \"black\" if (group[\"train_test\"] < 0).all() else \"blue\"\n",
    "    lw = 4.0\n",
    "    ax.plot(group.index, group.Retorno_Modelo_Acumulado, c=color, linewidth=lw)\n",
    "    plt.xlabel(\"Ano\"\n",
    "           , fontsize = 18);\n",
    "    plt.ylabel(\"Retorno Acumulado (%)\"\n",
    "           , fontsize = 18);\n",
    "    plt.title(\"K-Means - 7 anos no DOLFUT\", fontsize = 24)\n",
    "    plt.axvline( x = end_train, color='purple', linestyle = \"--\", lw = 2)\n",
    "    setup = dict(size = 16, color = \"darkgreen\")\n",
    "    setup2 = dict(size = 14, color = \"black\")\n",
    "    plt.text(\"2009-08-01\", 80, \"Treinamento: 2005-2012\", **setup)\n",
    "    plt.text(\"2017-01-01\", 70, \"Teste: 2013-2020\", **setup)\n",
    "    plt.text(\"2010-01-01\", 76.5, \"Retorno total: +108%\", **setup2)\n",
    "    plt.text(\"2010-01-01\", 73.0, \"Retorno teste: +56%\", **setup2)\n",
    "    plt.text(\"2017-03-01\", 75.5, \"www.outspokenmarket.com\", **setup2)\n",
    "\n",
    "base_agregada.groupby((base_agregada[\"train_test\"].shift() * base_agregada[\"train_test\"] < 0).cumsum()).apply(grafico)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "test_clusters.groupby(\"Cluster\")[\"Alvo1\"].sum()"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}