code/notebooks/correlaciones.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "from scipy.stats import shapiro, pointbiserialr\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "from scipy.stats import chi2_contingency\n",
    "import os\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [],
   "source": [
    "dataset_dir = os.path.join(os.path.dirname(os.path.dirname(os.path.abspath(''))), 'datasets')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "metadata": {},
   "outputs": [],
   "source": [
    "llamados= pd.read_excel(os.path.join(dataset_dir, 'xlsx/llamados_v2.xlsx'), parse_dates=['llamado_fecha_hora'])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['llamado_fecha_hora', 'llamado_provincia', 'llamante_edad',\n",
       "       'llamante_genero', 'llamante_vinculo', 'caso_judicializado',\n",
       "       'hecho_lugar', 'victima_a_resguardo', 'victima_edad', 'victima_genero',\n",
       "       'victima_nacionalidad', 'victima_vinculo_agresor',\n",
       "       'victima_discapacidad', 'victima_convive_agresor',\n",
       "       'vs_violacion_via_vaginal', 'vs_violacion_via_anal',\n",
       "       'vs_violacion_via_oral', 'vs_tentativa_violacion',\n",
       "       'vs_tocamiento_sexual', 'vs_intento_tocamiento',\n",
       "       'vs_Intento_violación_tercera_persona', 'vs_grooming',\n",
       "       'vs_exhibicionismo', 'vs_amenazas_verbales_contenido_sexual',\n",
       "       'vs_explotacion_sexual', 'vs_explotacion_sexual_comercial',\n",
       "       'vs_explotacion_sexual_viajes_turismo',\n",
       "       'vs_sospecha_trata_personas_fines_sexuales',\n",
       "       'vs_existencia_facilitador_corrupcion_nnya',\n",
       "       'vs_obligacion_sacarse_fotos_pornograficas',\n",
       "       'vs_eyaculacion_partes_cuerpo', 'vs_acoso_sexual',\n",
       "       'vs_iniciacion_sexual_forzada_inducida',\n",
       "       'vs_otra_forma_violencia_sexual', 'vs_no_sabe_no_contesta',\n",
       "       'ofv_sentimiento_amenaza', 'ofv_amenaza_explicita',\n",
       "       'ofv_violencia_fisica', 'ofv_intento_ahorcar', 'ofv_intento_quemar',\n",
       "       'ofv_intento_ahogar', 'ofv_amenaza_muerte',\n",
       "       'ofv_uso_sustancias_psicoactivas', 'ofv_intento_privacion_libertad',\n",
       "       'ofv_privacion_libertad', 'ofv_uso_arma_blanca', 'ofv_uso_arma_fuego',\n",
       "       'ofv_enganio_seduccion', 'ofv_intento_matar',\n",
       "       'ofv_uso_animal_victimizar', 'ofv_grooming', 'ofv_otra_forma_violencia',\n",
       "       'ofv_no_sabe_no_contesta'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 81,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "llamados.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "metadata": {},
   "outputs": [],
   "source": [
    "llamados_3= pd.read_excel(os.path.join(dataset_dir, 'xlsx/llamados_v3.xlsx'), parse_dates=['llamado_fecha_hora'])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "method"
      ]
     },
     "execution_count": 91,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "a = llamados.columns.tolist\n",
    "b = llamados_3.columns.tolist\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['llamado_fecha_hora', 'llamante_edad', 'llamante_genero',\n",
       "       'caso_judicializado', 'victima_a_resguardo', 'victima_edad',\n",
       "       'victima_genero', 'victima_discapacidad', 'victima_convive_agresor',\n",
       "       'vs_tocamiento_sexual', 'vs_intento_tocamiento', 'vs_grooming',\n",
       "       'vs_exhibicionismo', 'vs_obligacion_sacarse_fotos_pornograficas',\n",
       "       'vs_acoso_sexual', 'vs_iniciacion_sexual_forzada_inducida',\n",
       "       'vs_otra_forma_violencia_sexual', 'vs_no_sabe_no_contesta',\n",
       "       'ofv_sentimiento_amenaza', 'ofv_amenaza_explicita',\n",
       "       'ofv_violencia_fisica', 'ofv_enganio_seduccion', 'ofv_grooming',\n",
       "       'ofv_otra_forma_violencia', 'ofv_no_sabe_no_contesta',\n",
       "       'agresor_conocido_no_conocido_red', 'vinculo_llamante_red',\n",
       "       'hecho_lugar_red', 'llamado_provincia_red', 'victima_nacionalidad_red',\n",
       "       'fin_de_semana', 'momento_dia', 'estacion_del_año',\n",
       "       'vs_explotacion_sexual_group', 'vs_violacion_group',\n",
       "       'vs_tentativa_group'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 84,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "llamados_3.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Se eliminaron las columnas poco informativas:  ['vs_Intento_violación_tercera_persona', 'vs_amenazas_verbales_contenido_sexual', 'vs_explotacion_sexual_comercial', 'vs_explotacion_sexual_viajes_turismo', 'vs_sospecha_trata_personas_fines_sexuales', 'vs_existencia_facilitador_corrupcion_nnya', 'vs_eyaculacion_partes_cuerpo', 'ofv_intento_ahorcar', 'ofv_intento_quemar', 'ofv_intento_ahogar', 'ofv_amenaza_muerte', 'ofv_uso_sustancias_psicoactivas', 'ofv_intento_privacion_libertad', 'ofv_privacion_libertad', 'ofv_uso_arma_blanca', 'ofv_uso_arma_fuego', 'ofv_intento_matar', 'ofv_uso_animal_victimizar']\n"
     ]
    }
   ],
   "source": [
    "columnas_pocos_si = []\n",
    "for i in llamados:\n",
    "    if (llamados[i]== 'SI').sum() < 191:\n",
    "        columnas_pocos_si.append(i)\n",
    "    \n",
    "vs = list(llamados.loc[:, llamados.columns.str.startswith('vs')].columns)\n",
    "ofv = list(llamados.loc[:, llamados.columns.str.startswith('ofv')].columns)\n",
    "borradas = []\n",
    "for i in vs:\n",
    "    if i in columnas_pocos_si:\n",
    "        borradas.append(i)\n",
    "        llamados.drop(i, axis=1, inplace=True)\n",
    "\n",
    "for i in ofv:\n",
    "    if i in columnas_pocos_si:\n",
    "        borradas.append(i)\n",
    "        llamados.drop(i, axis=1, inplace=True)   \n",
    "\n",
    "print('Se eliminaron las columnas poco informativas: ', borradas)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "28"
      ]
     },
     "execution_count": 82,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(columnas_pocos_si)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['llamado_fecha_hora', 'llamado_provincia', 'llamante_edad',\n",
       "       'llamante_genero', 'llamante_vinculo', 'caso_judicializado',\n",
       "       'hecho_lugar', 'victima_a_resguardo', 'victima_edad', 'victima_genero',\n",
       "       'victima_nacionalidad', 'victima_vinculo_agresor',\n",
       "       'victima_discapacidad', 'victima_convive_agresor',\n",
       "       'vs_violacion_via_vaginal', 'vs_violacion_via_anal',\n",
       "       'vs_violacion_via_oral', 'vs_tentativa_violacion',\n",
       "       'vs_tocamiento_sexual', 'vs_intento_tocamiento', 'vs_grooming',\n",
       "       'vs_exhibicionismo', 'vs_explotacion_sexual',\n",
       "       'vs_obligacion_sacarse_fotos_pornograficas', 'vs_acoso_sexual',\n",
       "       'vs_iniciacion_sexual_forzada_inducida',\n",
       "       'vs_otra_forma_violencia_sexual', 'vs_no_sabe_no_contesta',\n",
       "       'ofv_sentimiento_amenaza', 'ofv_amenaza_explicita',\n",
       "       'ofv_violencia_fisica', 'ofv_enganio_seduccion', 'ofv_grooming',\n",
       "       'ofv_otra_forma_violencia', 'ofv_no_sabe_no_contesta'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 79,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "llamados.columns"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Point biserial correlation \n",
    "Para medir asociación entre variables binarias y continuas"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [],
   "source": [
    "completo_edades = llamados[~(llamados['victima_edad'].isnull()) & ~(llamados['llamante_edad'].isnull()) & ~(llamados['victima_convive_agresor'] == 'NS/NC')]\n",
    "\n",
    "\n",
    "del llamados"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Map target to binary values\n",
    "completo_edades['victima_convive_agresor'] = completo_edades['victima_convive_agresor'].map({'SI': 1, 'NO': 0})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Shapiro-Wilk Test: Statistics=0.886, p=0.000\n",
      "Sample does not look Gaussian (reject H0)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/var/folders/gm/4r6lhs5j5p12g3_ml4zjfy500000gq/T/ipykernel_68569/1712167121.py:2: UserWarning: scipy.stats.shapiro: For N > 5000, computed p-value may not be accurate. Current N is 8936.\n",
      "  stat, p = shapiro(completo_edades['victima_edad'])\n"
     ]
    }
   ],
   "source": [
    "# Check normality of continuous variable\n",
    "stat, p = shapiro(completo_edades['victima_edad'])\n",
    "print('Shapiro-Wilk Test: Statistics=%.3f, p=%.3f' % (stat, p))\n",
    "if p > 0.05:\n",
    "    print('Sample looks Gaussian (fail to reject H0)')\n",
    "else:\n",
    "    print('Sample does not look Gaussian (reject H0)')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Shapiro-Wilk Test: Statistics=0.974, p=0.000\n",
      "Sample does not look Gaussian (reject H0)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/var/folders/gm/4r6lhs5j5p12g3_ml4zjfy500000gq/T/ipykernel_68569/3329373257.py:2: UserWarning: scipy.stats.shapiro: For N > 5000, computed p-value may not be accurate. Current N is 8936.\n",
      "  stat, p = shapiro(completo_edades['llamante_edad'])\n"
     ]
    }
   ],
   "source": [
    "# Check normality of continuous variable\n",
    "stat, p = shapiro(completo_edades['llamante_edad'])\n",
    "print('Shapiro-Wilk Test: Statistics=%.3f, p=%.3f' % (stat, p))\n",
    "if p > 0.05:\n",
    "    print('Sample looks Gaussian (fail to reject H0)')\n",
    "else:\n",
    "    print('Sample does not look Gaussian (reject H0)')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Visual inspection\n",
    "sns.histplot(completo_edades['victima_edad'], kde=True)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Visual inspection\n",
    "sns.histplot(completo_edades['llamante_edad'], kde=True)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [],
   "source": [
    "# point biserial correlation\n",
    "from scipy.stats import  pointbiserialr\n",
    "\n",
    "numerical_variables = ['llamante_edad','victima_edad']\n",
    "\n",
    "# Calculate Point-Biserial correlation for numerical and binary categorical variables\n",
    "point_biserial_results = {}\n",
    "for variable in numerical_variables:\n",
    "        correlation, p = pointbiserialr(completo_edades[variable], completo_edades['victima_convive_agresor'])\n",
    "        point_biserial_results[variable] = {'Correlation': correlation, 'p-value': p}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "             llamante_edad  victima_edad\n",
      "Correlation         0.0377       -0.2005\n",
      "p-value             0.0004        0.0000\n"
     ]
    }
   ],
   "source": [
    "# Convert results to DataFrame\n",
    "results_df = pd.DataFrame(point_biserial_results)\n",
    "print(results_df.round(4))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [],
   "source": [
    "del completo_edades"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Phi coefficient\n",
    "Para medir asociación entre variables binarias:violencia (todas), fin de semana,  resguardo, caso judicializado, contra convive. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [],
   "source": [
    "llamados= pd.read_excel(os.path.join(dataset_dir, 'xlsx/llamados_v2.xlsx'), parse_dates=['llamado_fecha_hora'])\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Se eliminaron las columnas poco informativas:  ['vs_Intento_violación_tercera_persona', 'vs_amenazas_verbales_contenido_sexual', 'vs_explotacion_sexual_comercial', 'vs_explotacion_sexual_viajes_turismo', 'vs_sospecha_trata_personas_fines_sexuales', 'vs_existencia_facilitador_corrupcion_nnya', 'vs_eyaculacion_partes_cuerpo', 'ofv_intento_ahorcar', 'ofv_intento_quemar', 'ofv_intento_ahogar', 'ofv_amenaza_muerte', 'ofv_uso_sustancias_psicoactivas', 'ofv_intento_privacion_libertad', 'ofv_privacion_libertad', 'ofv_uso_arma_blanca', 'ofv_uso_arma_fuego', 'ofv_intento_matar', 'ofv_uso_animal_victimizar']\n"
     ]
    }
   ],
   "source": [
    "columnas_pocos_si = []\n",
    "for i in llamados:\n",
    "    if (llamados[i]== 'SI').sum() < 191:\n",
    "        columnas_pocos_si.append(i)\n",
    "    \n",
    "vs = list(llamados.loc[:, llamados.columns.str.startswith('vs')].columns)\n",
    "ofv = list(llamados.loc[:, llamados.columns.str.startswith('ofv')].columns)\n",
    "borradas = []\n",
    "for i in vs:\n",
    "    if i in columnas_pocos_si:\n",
    "        borradas.append(i)\n",
    "        llamados.drop(i, axis=1, inplace=True)\n",
    "\n",
    "for i in ofv:\n",
    "    if i in columnas_pocos_si:\n",
    "        borradas.append(i)\n",
    "        llamados.drop(i, axis=1, inplace=True)   \n",
    "\n",
    "print('Se eliminaron las columnas poco informativas: ', borradas)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {},
   "outputs": [],
   "source": [
    "cat_list= []\n",
    "for col in llamados.columns:\n",
    "    cat_list.append(col)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [],
   "source": [
    "vs = list(llamados.loc[:, llamados.columns.str.startswith('vs')].columns)\n",
    "ofv = list(llamados.loc[:, llamados.columns.str.startswith('ofv')].columns)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['vs_violacion_via_vaginal',\n",
       " 'vs_violacion_via_anal',\n",
       " 'vs_violacion_via_oral',\n",
       " 'vs_tentativa_violacion',\n",
       " 'vs_tocamiento_sexual',\n",
       " 'vs_intento_tocamiento',\n",
       " 'vs_grooming',\n",
       " 'vs_exhibicionismo',\n",
       " 'vs_explotacion_sexual',\n",
       " 'vs_obligacion_sacarse_fotos_pornograficas',\n",
       " 'vs_acoso_sexual',\n",
       " 'vs_iniciacion_sexual_forzada_inducida',\n",
       " 'vs_otra_forma_violencia_sexual',\n",
       " 'vs_no_sabe_no_contesta',\n",
       " 'ofv_sentimiento_amenaza',\n",
       " 'ofv_amenaza_explicita',\n",
       " 'ofv_violencia_fisica',\n",
       " 'ofv_enganio_seduccion',\n",
       " 'ofv_grooming',\n",
       " 'ofv_otra_forma_violencia',\n",
       " 'ofv_no_sabe_no_contesta',\n",
       " 'caso_judicializado',\n",
       " 'victima_a_resguardo']"
      ]
     },
     "execution_count": 66,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "bin_list= vs + ofv\n",
    "bin_list.append('caso_judicializado')\n",
    "bin_list.append('victima_a_resguardo')\n",
    "bin_list"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {},
   "outputs": [],
   "source": [
    "completo_convive = llamados[~(llamados['victima_convive_agresor'] == 'NS/NC')]\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {},
   "outputs": [],
   "source": [
    "del llamados"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Map target to binary values\n",
    "completo_convive.loc[:,'victima_convive_agresor'] = completo_convive.loc[:,'victima_convive_agresor'].map({'SI': 1, 'NO': 0})\n",
    "\n",
    "for col in bin_list:\n",
    "    completo_convive[col] = completo_convive[col].map({'SI': 1, 'NO':0})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Calculate phi coefficient\n",
    "def calculate_phi_coefficient(contingency_table):\n",
    "    chi2, p, dof, expected = chi2_contingency(contingency_table)\n",
    "    n = contingency_table.sum().sum()\n",
    "    phi = np.sqrt(chi2 / n)\n",
    "    return phi, p\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                                     Variable  Phi Coefficient  P-value\n",
      "0                    vs_violacion_via_vaginal            0.002    0.789\n",
      "1                       vs_violacion_via_anal            0.013    0.111\n",
      "2                       vs_violacion_via_oral            0.018    0.030\n",
      "3                      vs_tentativa_violacion            0.006    0.463\n",
      "4                        vs_tocamiento_sexual            0.016    0.049\n",
      "5                       vs_intento_tocamiento            0.004    0.655\n",
      "6                                 vs_grooming            0.165    0.000\n",
      "7                           vs_exhibicionismo            0.025    0.002\n",
      "8                       vs_explotacion_sexual            0.049    0.000\n",
      "9   vs_obligacion_sacarse_fotos_pornograficas            0.015    0.063\n",
      "10                            vs_acoso_sexual            0.035    0.000\n",
      "11      vs_iniciacion_sexual_forzada_inducida            0.023    0.005\n",
      "12             vs_otra_forma_violencia_sexual            0.025    0.002\n",
      "13                     vs_no_sabe_no_contesta            0.111    0.000\n",
      "14                    ofv_sentimiento_amenaza            0.066    0.000\n",
      "15                      ofv_amenaza_explicita            0.014    0.085\n",
      "16                       ofv_violencia_fisica            0.041    0.000\n",
      "17                      ofv_enganio_seduccion            0.036    0.000\n",
      "18                               ofv_grooming            0.066    0.000\n",
      "19                   ofv_otra_forma_violencia            0.002    0.797\n",
      "20                    ofv_no_sabe_no_contesta            0.096    0.000\n",
      "21                         caso_judicializado            0.015    0.072\n",
      "22                        victima_a_resguardo            0.418    0.000\n"
     ]
    }
   ],
   "source": [
    "# List to store results\n",
    "phi_results = []\n",
    "\n",
    "# Iterate over each binary variable in bin_list\n",
    "for var in bin_list:\n",
    "    # Create a contingency table\n",
    "    contingency_table = pd.crosstab(completo_convive['victima_convive_agresor'], completo_convive[var])\n",
    "    \n",
    "    # Calculate phi coefficient\n",
    "    phi, p_value = calculate_phi_coefficient(contingency_table)\n",
    "    \n",
    "    # Append results\n",
    "    phi_results.append({'Variable': var, 'Phi Coefficient': round(phi, 3), 'P-value': round(p_value, 3)})\n",
    "\n",
    "# Convert results to DataFrame\n",
    "phi_results_df = pd.DataFrame(phi_results)\n",
    "\n",
    "# Print the results\n",
    "print(phi_results_df)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Correlación Chi2 "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {},
   "outputs": [],
   "source": [
    "chi_list = [n for n in cat_list if n not in bin_list]\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Chequeamos que se cumplan las condiciones previas"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Condition 5: Not all expected frequencies are greater than 1.\n",
      "Condition 6: Proportion of cells with expected frequencies less than 5 is not acceptable.\n"
     ]
    }
   ],
   "source": [
    "# Initialize a list to store contingency tables\n",
    "contingency_tables = []\n",
    "\n",
    "# Iterate through each categorical variable\n",
    "for variable in chi_list:\n",
    "    # Create a contingency table\n",
    "    contingency_table = pd.crosstab(index=completo_convive['victima_convive_agresor'], columns=completo_convive[variable])\n",
    "    \n",
    "    # Append the contingency table to the list\n",
    "    contingency_tables.append(contingency_table)\n",
    "\n",
    "# Combine all contingency tables into a single table\n",
    "combined_contingency_table = pd.concat(contingency_tables, axis=1)\n",
    "\n",
    "# Calculate expected frequencies\n",
    "expected_frequencies = chi2_contingency(combined_contingency_table)[3]\n",
    "\n",
    "# Check condition 5: No expected frequencies should be 0\n",
    "if (expected_frequencies > 1).all():\n",
    "    print(\"Condition 5: All expected frequencies are greater than 1.\")\n",
    "else:\n",
    "    print(\"Condition 5: Not all expected frequencies are greater than 1.\")\n",
    "\n",
    "# Check condition 6: Proportion of cells with expected frequencies less than 5\n",
    "cells_less_than_5 = (expected_frequencies < 5).sum()\n",
    "total_cells = combined_contingency_table.size\n",
    "proportion_less_than_5 = cells_less_than_5 / total_cells\n",
    "\n",
    "if proportion_less_than_5 <= 0.2:\n",
    "    print(\"Condition 6: Proportion of cells with expected frequencies less than 5 is acceptable.\")\n",
    "else:\n",
    "    print(\"Condition 6: Proportion of cells with expected frequencies less than 5 is not acceptable.\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "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>Variable</th>\n",
       "      <th>Chi-Squared Statistic</th>\n",
       "      <th>P-value</th>\n",
       "      <th>DoF</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>llamado_fecha_hora</td>\n",
       "      <td>14545.727994</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>13411</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>llamado_provincia</td>\n",
       "      <td>88.194934</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>llamante_edad</td>\n",
       "      <td>233.435050</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>75</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>llamante_genero</td>\n",
       "      <td>19.099752</td>\n",
       "      <td>0.0003</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>llamante_vinculo</td>\n",
       "      <td>1578.200948</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>hecho_lugar</td>\n",
       "      <td>2913.856872</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>16</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>victima_edad</td>\n",
       "      <td>734.511120</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>85</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>victima_genero</td>\n",
       "      <td>29.437760</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>victima_nacionalidad</td>\n",
       "      <td>79.793899</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>victima_vinculo_agresor</td>\n",
       "      <td>3950.050272</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>victima_discapacidad</td>\n",
       "      <td>137.850886</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>victima_convive_agresor</td>\n",
       "      <td>15078.303693</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                   Variable  Chi-Squared Statistic  P-value    DoF\n",
       "0        llamado_fecha_hora           14545.727994   0.0000  13411\n",
       "1         llamado_provincia              88.194934   0.0000     24\n",
       "2             llamante_edad             233.435050   0.0000     75\n",
       "3           llamante_genero              19.099752   0.0003      3\n",
       "4          llamante_vinculo            1578.200948   0.0000     14\n",
       "5               hecho_lugar            2913.856872   0.0000     16\n",
       "6              victima_edad             734.511120   0.0000     85\n",
       "7            victima_genero              29.437760   0.0000      3\n",
       "8      victima_nacionalidad              79.793899   0.0000      8\n",
       "9   victima_vinculo_agresor            3950.050272   0.0000     14\n",
       "10     victima_discapacidad             137.850886   0.0000      2\n",
       "11  victima_convive_agresor           15078.303693   0.0000      1"
      ]
     },
     "execution_count": 74,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Initialize an empty list to store results\n",
    "results = []\n",
    "\n",
    "\n",
    "# Perform chi-squared test for each variable combination\n",
    "for variable in chi_list:\n",
    "    # Create a contingency table\n",
    "    contingency_table = pd.crosstab(index=completo_convive['victima_convive_agresor'], columns=completo_convive[variable])\n",
    "    \n",
    "    # Perform chi-squared test\n",
    "    chi2, p, dof, expected = chi2_contingency(contingency_table)\n",
    "    \n",
    "    # Round p-value to 4 decimal places\n",
    "    p_rounded = round(p, 4)\n",
    "\n",
    "    # Append results to the list\n",
    "    results.append({'Variable': variable, 'Chi-Squared Statistic': chi2, 'P-value': p_rounded, 'DoF': dof})\n",
    "\n",
    "# Convert results list to DataFrame\n",
    "results_df = pd.DataFrame(results)\n",
    "\n",
    "# Print the resulting table\n",
    "results_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "llamante_genero\n",
       "Femenino       10782\n",
       "Masculino       4174\n",
       "NS/NC            116\n",
       "Transgénero       13\n",
       "Name: count, dtype: int64"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "completo_convive.llamante_genero.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [],
   "source": [
    "contingency_table = pd.crosstab(index=completo_convive['victima_convive_agresor'], columns=completo_convive['llamante_genero'])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "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>llamante_genero</th>\n",
       "      <th>Femenino</th>\n",
       "      <th>Masculino</th>\n",
       "      <th>NS/NC</th>\n",
       "      <th>Transgénero</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>victima_convive_agresor</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>8851</td>\n",
       "      <td>3388</td>\n",
       "      <td>78</td>\n",
       "      <td>12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1931</td>\n",
       "      <td>786</td>\n",
       "      <td>38</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "llamante_genero          Femenino  Masculino  NS/NC  Transgénero\n",
       "victima_convive_agresor                                         \n",
       "0                            8851       3388     78           12\n",
       "1                            1931        786     38            1"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "contingency_table"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "tpenv",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}