06-pair-gestion-nulos.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "# Imputación de nulos usando métodos avanzados estadísticos\n",
    "# -----------------------------------------------------------------------\n",
    "from sklearn.impute import SimpleImputer\n",
    "from sklearn.experimental import enable_iterative_imputer\n",
    "from sklearn.impute import IterativeImputer\n",
    "from sklearn.impute import KNNImputer\n",
    "\n",
    "# Librerías de visualización\n",
    "# -----------------------------------------------------------------------\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n",
    "# Configuración\n",
    "# -----------------------------------------------------------------------\n",
    "pd.set_option('display.max_columns', None) # para poder visualizar todas las columnas de los DataFrames"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Instrucciones:\n",
    "\n",
    "1- Identificación de Valores Nulos: Identifica todas las columnas que contengan valores nulos en el DataFrame.\n",
    "    \n",
    "2- Selección de Método de Imputación: Discute en tu compañera cuál sería la mejor estrategia para manejar los valores nulos en cada una de las columnas identificadas en el paso anterior.\n",
    "\n",
    "3- Imputación de Valores Nulos: Implementa el método de imputación seleccionado en el paso 2 para llenar los valores nulos en las columnas.\n",
    "\n",
    "4- Informe: Añade al final de un jupyter una explicación breve que describa las columnas que tenían valores nulos, cómo decidiste imputarlos y cualquier observación adicional que consideres importante sobre el proceso de limpieza de datos.\n",
    "\n",
    "Nota: Puedes utilizar cualquier método o estrategia de imputación que consideres adecuado para los datos y discutir las ventajas y desventajas de tu elección en el inform"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_world = pd.read_csv(\"merge_world_2.csv\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(195, 36)"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_world.shape"
   ]
  },
  {
   "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>country</th>\n",
       "      <th>density</th>\n",
       "      <th>abbreviation</th>\n",
       "      <th>agriculturalland</th>\n",
       "      <th>landarea</th>\n",
       "      <th>armedforcessize</th>\n",
       "      <th>birthrate</th>\n",
       "      <th>callingcode</th>\n",
       "      <th>capital/majorcity</th>\n",
       "      <th>co2-emissions</th>\n",
       "      <th>cpi</th>\n",
       "      <th>cpichange</th>\n",
       "      <th>currency-code</th>\n",
       "      <th>fertilityrate</th>\n",
       "      <th>forestedarea</th>\n",
       "      <th>gasolineprice</th>\n",
       "      <th>gdp</th>\n",
       "      <th>grossprimaryeducationenrollment</th>\n",
       "      <th>grosstertiaryeducationenrollment</th>\n",
       "      <th>infantmortality</th>\n",
       "      <th>largestcity</th>\n",
       "      <th>lifeexpectancy</th>\n",
       "      <th>maternalmortalityratio</th>\n",
       "      <th>minimumwage</th>\n",
       "      <th>officiallanguage</th>\n",
       "      <th>outofpockethealthexpenditure</th>\n",
       "      <th>physiciansperthousand</th>\n",
       "      <th>population</th>\n",
       "      <th>populationlaborforceparticipation</th>\n",
       "      <th>taxrevenue</th>\n",
       "      <th>totaltaxrate</th>\n",
       "      <th>unemploymentrate</th>\n",
       "      <th>urban_population</th>\n",
       "      <th>longitud</th>\n",
       "      <th>latitud</th>\n",
       "      <th>continent</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Afghanistan</td>\n",
       "      <td>60</td>\n",
       "      <td>AF</td>\n",
       "      <td>58.1</td>\n",
       "      <td>652230.0</td>\n",
       "      <td>323000.0</td>\n",
       "      <td>32.49</td>\n",
       "      <td>93.0</td>\n",
       "      <td>Kabul</td>\n",
       "      <td>8672.0</td>\n",
       "      <td>149.90</td>\n",
       "      <td>2.3</td>\n",
       "      <td>AFN</td>\n",
       "      <td>4.47</td>\n",
       "      <td>2.1</td>\n",
       "      <td>0.70</td>\n",
       "      <td>1.910135e+10</td>\n",
       "      <td>104.0</td>\n",
       "      <td>9.7</td>\n",
       "      <td>47.9</td>\n",
       "      <td>Kabul</td>\n",
       "      <td>64.5</td>\n",
       "      <td>638.0</td>\n",
       "      <td>0.43</td>\n",
       "      <td>Pashto</td>\n",
       "      <td>78.4</td>\n",
       "      <td>0.28</td>\n",
       "      <td>38041754.0</td>\n",
       "      <td>48.9</td>\n",
       "      <td>9.3</td>\n",
       "      <td>71.4</td>\n",
       "      <td>11.12</td>\n",
       "      <td>9797273.0</td>\n",
       "      <td>33.939110</td>\n",
       "      <td>67.709953</td>\n",
       "      <td>Asia</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Albania</td>\n",
       "      <td>105</td>\n",
       "      <td>AL</td>\n",
       "      <td>43.1</td>\n",
       "      <td>28748.0</td>\n",
       "      <td>9000.0</td>\n",
       "      <td>11.78</td>\n",
       "      <td>355.0</td>\n",
       "      <td>Tirana</td>\n",
       "      <td>4536.0</td>\n",
       "      <td>119.05</td>\n",
       "      <td>1.4</td>\n",
       "      <td>ALL</td>\n",
       "      <td>1.62</td>\n",
       "      <td>28.1</td>\n",
       "      <td>1.36</td>\n",
       "      <td>1.527808e+10</td>\n",
       "      <td>107.0</td>\n",
       "      <td>55.0</td>\n",
       "      <td>7.8</td>\n",
       "      <td>Tirana</td>\n",
       "      <td>78.5</td>\n",
       "      <td>15.0</td>\n",
       "      <td>1.12</td>\n",
       "      <td>Albanian</td>\n",
       "      <td>56.9</td>\n",
       "      <td>1.20</td>\n",
       "      <td>2854191.0</td>\n",
       "      <td>55.7</td>\n",
       "      <td>18.6</td>\n",
       "      <td>36.6</td>\n",
       "      <td>12.33</td>\n",
       "      <td>1747593.0</td>\n",
       "      <td>41.153332</td>\n",
       "      <td>20.168331</td>\n",
       "      <td>Europe</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Algeria</td>\n",
       "      <td>18</td>\n",
       "      <td>DZ</td>\n",
       "      <td>17.4</td>\n",
       "      <td>2381741.0</td>\n",
       "      <td>317000.0</td>\n",
       "      <td>24.28</td>\n",
       "      <td>213.0</td>\n",
       "      <td>Algiers</td>\n",
       "      <td>150006.0</td>\n",
       "      <td>151.36</td>\n",
       "      <td>2.0</td>\n",
       "      <td>DZD</td>\n",
       "      <td>3.02</td>\n",
       "      <td>0.8</td>\n",
       "      <td>0.28</td>\n",
       "      <td>1.699882e+11</td>\n",
       "      <td>109.9</td>\n",
       "      <td>51.4</td>\n",
       "      <td>20.1</td>\n",
       "      <td>Algiers</td>\n",
       "      <td>76.7</td>\n",
       "      <td>112.0</td>\n",
       "      <td>0.95</td>\n",
       "      <td>Arabic</td>\n",
       "      <td>28.1</td>\n",
       "      <td>1.72</td>\n",
       "      <td>43053054.0</td>\n",
       "      <td>41.2</td>\n",
       "      <td>37.2</td>\n",
       "      <td>66.1</td>\n",
       "      <td>11.70</td>\n",
       "      <td>31510100.0</td>\n",
       "      <td>28.033886</td>\n",
       "      <td>1.659626</td>\n",
       "      <td>Africa</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       country density abbreviation  agriculturalland   landarea  \\\n",
       "0  Afghanistan      60           AF              58.1   652230.0   \n",
       "1      Albania     105           AL              43.1    28748.0   \n",
       "2      Algeria      18           DZ              17.4  2381741.0   \n",
       "\n",
       "   armedforcessize  birthrate  callingcode capital/majorcity  co2-emissions  \\\n",
       "0         323000.0      32.49         93.0             Kabul         8672.0   \n",
       "1           9000.0      11.78        355.0            Tirana         4536.0   \n",
       "2         317000.0      24.28        213.0           Algiers       150006.0   \n",
       "\n",
       "      cpi  cpichange currency-code  fertilityrate  forestedarea  \\\n",
       "0  149.90        2.3           AFN           4.47           2.1   \n",
       "1  119.05        1.4           ALL           1.62          28.1   \n",
       "2  151.36        2.0           DZD           3.02           0.8   \n",
       "\n",
       "   gasolineprice           gdp  grossprimaryeducationenrollment  \\\n",
       "0           0.70  1.910135e+10                            104.0   \n",
       "1           1.36  1.527808e+10                            107.0   \n",
       "2           0.28  1.699882e+11                            109.9   \n",
       "\n",
       "   grosstertiaryeducationenrollment  infantmortality largestcity  \\\n",
       "0                               9.7             47.9       Kabul   \n",
       "1                              55.0              7.8      Tirana   \n",
       "2                              51.4             20.1     Algiers   \n",
       "\n",
       "   lifeexpectancy  maternalmortalityratio  minimumwage officiallanguage  \\\n",
       "0            64.5                   638.0         0.43           Pashto   \n",
       "1            78.5                    15.0         1.12         Albanian   \n",
       "2            76.7                   112.0         0.95           Arabic   \n",
       "\n",
       "   outofpockethealthexpenditure  physiciansperthousand  population  \\\n",
       "0                          78.4                   0.28  38041754.0   \n",
       "1                          56.9                   1.20   2854191.0   \n",
       "2                          28.1                   1.72  43053054.0   \n",
       "\n",
       "   populationlaborforceparticipation  taxrevenue  totaltaxrate  \\\n",
       "0                               48.9         9.3          71.4   \n",
       "1                               55.7        18.6          36.6   \n",
       "2                               41.2        37.2          66.1   \n",
       "\n",
       "   unemploymentrate  urban_population   longitud    latitud continent  \n",
       "0             11.12         9797273.0  33.939110  67.709953      Asia  \n",
       "1             12.33         1747593.0  41.153332  20.168331    Europe  \n",
       "2             11.70        31510100.0  28.033886   1.659626    Africa  "
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_world.head(3)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "2. ¿Qué columnas tienen nulos? "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "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>%nulos</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>abbreviation</th>\n",
       "      <td>3.589744</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>agriculturalland</th>\n",
       "      <td>3.589744</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>landarea</th>\n",
       "      <td>0.512821</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>armedforcessize</th>\n",
       "      <td>12.307692</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>birthrate</th>\n",
       "      <td>3.076923</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>callingcode</th>\n",
       "      <td>0.512821</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>capital/majorcity</th>\n",
       "      <td>1.538462</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>co2-emissions</th>\n",
       "      <td>3.589744</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>cpi</th>\n",
       "      <td>8.717949</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>cpichange</th>\n",
       "      <td>8.205128</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>currency-code</th>\n",
       "      <td>7.692308</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>fertilityrate</th>\n",
       "      <td>3.589744</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>forestedarea</th>\n",
       "      <td>3.589744</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>gasolineprice</th>\n",
       "      <td>10.256410</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>gdp</th>\n",
       "      <td>1.025641</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>grossprimaryeducationenrollment</th>\n",
       "      <td>3.589744</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>grosstertiaryeducationenrollment</th>\n",
       "      <td>6.153846</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>infantmortality</th>\n",
       "      <td>3.076923</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>largestcity</th>\n",
       "      <td>3.076923</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>lifeexpectancy</th>\n",
       "      <td>4.102564</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>maternalmortalityratio</th>\n",
       "      <td>7.179487</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>minimumwage</th>\n",
       "      <td>23.076923</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>officiallanguage</th>\n",
       "      <td>2.564103</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>outofpockethealthexpenditure</th>\n",
       "      <td>3.589744</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>physiciansperthousand</th>\n",
       "      <td>3.589744</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>population</th>\n",
       "      <td>0.512821</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>populationlaborforceparticipation</th>\n",
       "      <td>9.743590</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>taxrevenue</th>\n",
       "      <td>13.333333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>totaltaxrate</th>\n",
       "      <td>6.153846</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>unemploymentrate</th>\n",
       "      <td>9.743590</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>urban_population</th>\n",
       "      <td>2.564103</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>longitud</th>\n",
       "      <td>0.512821</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>latitud</th>\n",
       "      <td>0.512821</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>continent</th>\n",
       "      <td>7.692308</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                      %nulos\n",
       "abbreviation                        3.589744\n",
       "agriculturalland                    3.589744\n",
       "landarea                            0.512821\n",
       "armedforcessize                    12.307692\n",
       "birthrate                           3.076923\n",
       "callingcode                         0.512821\n",
       "capital/majorcity                   1.538462\n",
       "co2-emissions                       3.589744\n",
       "cpi                                 8.717949\n",
       "cpichange                           8.205128\n",
       "currency-code                       7.692308\n",
       "fertilityrate                       3.589744\n",
       "forestedarea                        3.589744\n",
       "gasolineprice                      10.256410\n",
       "gdp                                 1.025641\n",
       "grossprimaryeducationenrollment     3.589744\n",
       "grosstertiaryeducationenrollment    6.153846\n",
       "infantmortality                     3.076923\n",
       "largestcity                         3.076923\n",
       "lifeexpectancy                      4.102564\n",
       "maternalmortalityratio              7.179487\n",
       "minimumwage                        23.076923\n",
       "officiallanguage                    2.564103\n",
       "outofpockethealthexpenditure        3.589744\n",
       "physiciansperthousand               3.589744\n",
       "population                          0.512821\n",
       "populationlaborforceparticipation   9.743590\n",
       "taxrevenue                         13.333333\n",
       "totaltaxrate                        6.153846\n",
       "unemploymentrate                    9.743590\n",
       "urban_population                    2.564103\n",
       "longitud                            0.512821\n",
       "latitud                             0.512821\n",
       "continent                           7.692308"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# lo primero que vamos a hacer es calcular el porcentaje de nulos que tenemos para cada columna\n",
    "# y los pasamos a DataFrame\n",
    "df_nulos = pd.DataFrame((df_world.isnull().sum() / df_world.shape[0]) * 100, columns=[\"%nulos\"])\n",
    "\n",
    "# filtramos el DataFrame para quedarnos solo con aquellas columnas que tengan nulos\n",
    "df_nulos[df_nulos[\"%nulos\"] > 0]"
   ]
  },
  {
   "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>%nulos</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>minimumwage</th>\n",
       "      <td>23.076923</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>taxrevenue</th>\n",
       "      <td>13.333333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>armedforcessize</th>\n",
       "      <td>12.307692</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>gasolineprice</th>\n",
       "      <td>10.256410</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>unemploymentrate</th>\n",
       "      <td>9.743590</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>populationlaborforceparticipation</th>\n",
       "      <td>9.743590</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>cpi</th>\n",
       "      <td>8.717949</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>cpichange</th>\n",
       "      <td>8.205128</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>continent</th>\n",
       "      <td>7.692308</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>currency-code</th>\n",
       "      <td>7.692308</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>maternalmortalityratio</th>\n",
       "      <td>7.179487</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>totaltaxrate</th>\n",
       "      <td>6.153846</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>grosstertiaryeducationenrollment</th>\n",
       "      <td>6.153846</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>lifeexpectancy</th>\n",
       "      <td>4.102564</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>fertilityrate</th>\n",
       "      <td>3.589744</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>physiciansperthousand</th>\n",
       "      <td>3.589744</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>co2-emissions</th>\n",
       "      <td>3.589744</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>grossprimaryeducationenrollment</th>\n",
       "      <td>3.589744</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>abbreviation</th>\n",
       "      <td>3.589744</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>agriculturalland</th>\n",
       "      <td>3.589744</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>forestedarea</th>\n",
       "      <td>3.589744</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>outofpockethealthexpenditure</th>\n",
       "      <td>3.589744</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>infantmortality</th>\n",
       "      <td>3.076923</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>largestcity</th>\n",
       "      <td>3.076923</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>birthrate</th>\n",
       "      <td>3.076923</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>urban_population</th>\n",
       "      <td>2.564103</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>officiallanguage</th>\n",
       "      <td>2.564103</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>capital/majorcity</th>\n",
       "      <td>1.538462</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>gdp</th>\n",
       "      <td>1.025641</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>population</th>\n",
       "      <td>0.512821</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>callingcode</th>\n",
       "      <td>0.512821</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>landarea</th>\n",
       "      <td>0.512821</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>longitud</th>\n",
       "      <td>0.512821</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>latitud</th>\n",
       "      <td>0.512821</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>density</th>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>country</th>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                      %nulos\n",
       "minimumwage                        23.076923\n",
       "taxrevenue                         13.333333\n",
       "armedforcessize                    12.307692\n",
       "gasolineprice                      10.256410\n",
       "unemploymentrate                    9.743590\n",
       "populationlaborforceparticipation   9.743590\n",
       "cpi                                 8.717949\n",
       "cpichange                           8.205128\n",
       "continent                           7.692308\n",
       "currency-code                       7.692308\n",
       "maternalmortalityratio              7.179487\n",
       "totaltaxrate                        6.153846\n",
       "grosstertiaryeducationenrollment    6.153846\n",
       "lifeexpectancy                      4.102564\n",
       "fertilityrate                       3.589744\n",
       "physiciansperthousand               3.589744\n",
       "co2-emissions                       3.589744\n",
       "grossprimaryeducationenrollment     3.589744\n",
       "abbreviation                        3.589744\n",
       "agriculturalland                    3.589744\n",
       "forestedarea                        3.589744\n",
       "outofpockethealthexpenditure        3.589744\n",
       "infantmortality                     3.076923\n",
       "largestcity                         3.076923\n",
       "birthrate                           3.076923\n",
       "urban_population                    2.564103\n",
       "officiallanguage                    2.564103\n",
       "capital/majorcity                   1.538462\n",
       "gdp                                 1.025641\n",
       "population                          0.512821\n",
       "callingcode                         0.512821\n",
       "landarea                            0.512821\n",
       "longitud                            0.512821\n",
       "latitud                             0.512821\n",
       "density                             0.000000\n",
       "country                             0.000000"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Ordenamos de mayor a menor por la columna '%nulos' para verlo mejor\n",
    "df_nulos_ordenado = df_nulos.sort_values(by='%nulos', ascending=False)\n",
    "df_nulos_ordenado"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "________\n",
    "# Primer análisis de los datos\n",
    "\n",
    "Tenemos 2 columnas que no tienen ningún dato nulo, por lo tanto en este proceso las podremos obviar en esta fase. Son las columnas:  density y country.\n",
    "\n",
    "- La columna con más porcentaje de nulos es \"mininum wage\", con un 23%. Decicimos mantenerla y gestionarla poniendo \"unknown\"\n",
    "- El resto de columnas van del 0,5 al 13% de nulos. A continuación gestionaremos las columnas caso a caso para determinar de qué manera gestionarlo. Primero miraremos los valores categóricos y después los numéricos.\n",
    "\n",
    "- Ésta decisión también vendrá  determinada por el tipo de datos con los que trabajamos y de la importáncia que éstos tengan en nuestro análisis. \n",
    "\n",
    "\n"
   ]
  },
  {
   "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>country</th>\n",
       "      <th>density</th>\n",
       "      <th>abbreviation</th>\n",
       "      <th>agriculturalland</th>\n",
       "      <th>landarea</th>\n",
       "      <th>armedforcessize</th>\n",
       "      <th>birthrate</th>\n",
       "      <th>callingcode</th>\n",
       "      <th>capital/majorcity</th>\n",
       "      <th>co2-emissions</th>\n",
       "      <th>cpi</th>\n",
       "      <th>cpichange</th>\n",
       "      <th>currency-code</th>\n",
       "      <th>fertilityrate</th>\n",
       "      <th>forestedarea</th>\n",
       "      <th>gasolineprice</th>\n",
       "      <th>gdp</th>\n",
       "      <th>grossprimaryeducationenrollment</th>\n",
       "      <th>grosstertiaryeducationenrollment</th>\n",
       "      <th>infantmortality</th>\n",
       "      <th>largestcity</th>\n",
       "      <th>lifeexpectancy</th>\n",
       "      <th>maternalmortalityratio</th>\n",
       "      <th>minimumwage</th>\n",
       "      <th>officiallanguage</th>\n",
       "      <th>outofpockethealthexpenditure</th>\n",
       "      <th>physiciansperthousand</th>\n",
       "      <th>population</th>\n",
       "      <th>populationlaborforceparticipation</th>\n",
       "      <th>taxrevenue</th>\n",
       "      <th>totaltaxrate</th>\n",
       "      <th>unemploymentrate</th>\n",
       "      <th>urban_population</th>\n",
       "      <th>longitud</th>\n",
       "      <th>latitud</th>\n",
       "      <th>continent</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Afghanistan</td>\n",
       "      <td>60</td>\n",
       "      <td>AF</td>\n",
       "      <td>58.1</td>\n",
       "      <td>652230.0</td>\n",
       "      <td>323000.0</td>\n",
       "      <td>32.49</td>\n",
       "      <td>93.0</td>\n",
       "      <td>Kabul</td>\n",
       "      <td>8672.0</td>\n",
       "      <td>149.9</td>\n",
       "      <td>2.3</td>\n",
       "      <td>AFN</td>\n",
       "      <td>4.47</td>\n",
       "      <td>2.1</td>\n",
       "      <td>0.7</td>\n",
       "      <td>1.910135e+10</td>\n",
       "      <td>104.0</td>\n",
       "      <td>9.7</td>\n",
       "      <td>47.9</td>\n",
       "      <td>Kabul</td>\n",
       "      <td>64.5</td>\n",
       "      <td>638.0</td>\n",
       "      <td>0.43</td>\n",
       "      <td>Pashto</td>\n",
       "      <td>78.4</td>\n",
       "      <td>0.28</td>\n",
       "      <td>38041754.0</td>\n",
       "      <td>48.9</td>\n",
       "      <td>9.3</td>\n",
       "      <td>71.4</td>\n",
       "      <td>11.12</td>\n",
       "      <td>9797273.0</td>\n",
       "      <td>33.93911</td>\n",
       "      <td>67.709953</td>\n",
       "      <td>Asia</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       country density abbreviation  agriculturalland  landarea  \\\n",
       "0  Afghanistan      60           AF              58.1  652230.0   \n",
       "\n",
       "   armedforcessize  birthrate  callingcode capital/majorcity  co2-emissions  \\\n",
       "0         323000.0      32.49         93.0             Kabul         8672.0   \n",
       "\n",
       "     cpi  cpichange currency-code  fertilityrate  forestedarea  gasolineprice  \\\n",
       "0  149.9        2.3           AFN           4.47           2.1            0.7   \n",
       "\n",
       "            gdp  grossprimaryeducationenrollment  \\\n",
       "0  1.910135e+10                            104.0   \n",
       "\n",
       "   grosstertiaryeducationenrollment  infantmortality largestcity  \\\n",
       "0                               9.7             47.9       Kabul   \n",
       "\n",
       "   lifeexpectancy  maternalmortalityratio  minimumwage officiallanguage  \\\n",
       "0            64.5                   638.0         0.43           Pashto   \n",
       "\n",
       "   outofpockethealthexpenditure  physiciansperthousand  population  \\\n",
       "0                          78.4                   0.28  38041754.0   \n",
       "\n",
       "   populationlaborforceparticipation  taxrevenue  totaltaxrate  \\\n",
       "0                               48.9         9.3          71.4   \n",
       "\n",
       "   unemploymentrate  urban_population  longitud    latitud continent  \n",
       "0             11.12         9797273.0  33.93911  67.709953      Asia  "
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_world.head(1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "_____________________________\n",
    "# Valores categóricos"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Las columnas categóricas que tienen nulos son : \n",
      " \n",
      "Index(['abbreviation', 'capital/majorcity', 'currency-code', 'largestcity',\n",
      "       'officiallanguage', 'continent'],\n",
      "      dtype='object')\n"
     ]
    }
   ],
   "source": [
    "# Obtenemos la lista de columnas categóricas que tienen nulos\n",
    "nulos_esta_cat = df_world[df_world.columns[df_world.isnull().any()]].select_dtypes(include = \"O\").columns\n",
    "print(\"Las columnas categóricas que tienen nulos son : \\n \")\n",
    "print(nulos_esta_cat)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "La distribución de las categorías para la columna ABBREVIATION\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "abbreviation\n",
       "AF    0.005128\n",
       "PY    0.005128\n",
       "NE    0.005128\n",
       "NG    0.005128\n",
       "KP    0.005128\n",
       "        ...   \n",
       "GR    0.005128\n",
       "GD    0.005128\n",
       "GT    0.005128\n",
       "GN    0.005128\n",
       "ZW    0.005128\n",
       "Name: count, Length: 188, dtype: float64"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "........................\n",
      "La distribución de las categorías para la columna CAPITAL/MAJORCITY\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "capital/majorcity\n",
       "Kabul                    0.005128\n",
       "Tirana                   0.005128\n",
       "Wellington               0.005128\n",
       "Managua                  0.005128\n",
       "Niamey                   0.005128\n",
       "                           ...   \n",
       "Athens                   0.005128\n",
       "St. George's, Grenada    0.005128\n",
       "Guatemala City           0.005128\n",
       "Conakry                  0.005128\n",
       "Harare                   0.005128\n",
       "Name: count, Length: 192, dtype: float64"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "........................\n",
      "La distribución de las categorías para la columna CURRENCY-CODE\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "currency-code\n",
       "EUR    0.117949\n",
       "XOF    0.041026\n",
       "XCD    0.030769\n",
       "USD    0.030769\n",
       "XAF    0.025641\n",
       "         ...   \n",
       "GMD    0.005128\n",
       "FJD    0.005128\n",
       "ETB    0.005128\n",
       "ERN    0.005128\n",
       "ZMW    0.005128\n",
       "Name: count, Length: 133, dtype: float64"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "........................\n",
      "La distribución de las categorías para la columna LARGESTCITY\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "largestcity\n",
       "S����                    0.010256\n",
       "Panama City              0.005128\n",
       "Managua                  0.005128\n",
       "Niamey                   0.005128\n",
       "Lagos                    0.005128\n",
       "                           ...   \n",
       "Macedonia                0.005128\n",
       "St. George's, Grenada    0.005128\n",
       "Guatemala City           0.005128\n",
       "Kankan                   0.005128\n",
       "Harare                   0.005128\n",
       "Name: count, Length: 188, dtype: float64"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "........................\n",
      "La distribución de las categorías para la columna OFFICIALLANGUAGE\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "officiallanguage\n",
       "English             0.158974\n",
       "French              0.128205\n",
       "Spanish             0.097436\n",
       "Arabic              0.092308\n",
       "Portuguese          0.035897\n",
       "                      ...   \n",
       "Albanian            0.005128\n",
       "Jamaican English    0.005128\n",
       "Lao                 0.005128\n",
       "Latvian             0.005128\n",
       "Shona               0.005128\n",
       "Name: count, Length: 76, dtype: float64"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "........................\n",
      "La distribución de las categorías para la columna CONTINENT\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "continent\n",
       "Africa           0.271795\n",
       "Asia             0.235897\n",
       "Europe           0.225641\n",
       "Oceania          0.066667\n",
       "South America    0.061538\n",
       "North America    0.061538\n",
       "Name: count, dtype: float64"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "........................\n"
     ]
    }
   ],
   "source": [
    "# sacamos el 'value_counts()' de cada una de las columnas categóricas que tienen nulos para saber como es la distribución de sus categorías\n",
    "for col in nulos_esta_cat:\n",
    "    print(f\"La distribución de las categorías para la columna {col.upper()}\")\n",
    "    display(df_world[col].value_counts() / df_world.shape[0])\n",
    "    print(\"........................\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "stats_descriptivas = pd.DataFrame()\n",
    "\n",
    "for x in nulos_esta_cat:\n",
    "    stats = df_world[x].describe()\n",
    "    stats_descriptivas[x] = stats"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "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>abbreviation</th>\n",
       "      <th>capital/majorcity</th>\n",
       "      <th>currency-code</th>\n",
       "      <th>largestcity</th>\n",
       "      <th>officiallanguage</th>\n",
       "      <th>continent</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>188</td>\n",
       "      <td>192</td>\n",
       "      <td>180</td>\n",
       "      <td>189</td>\n",
       "      <td>190</td>\n",
       "      <td>180</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>unique</th>\n",
       "      <td>188</td>\n",
       "      <td>192</td>\n",
       "      <td>133</td>\n",
       "      <td>188</td>\n",
       "      <td>76</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>top</th>\n",
       "      <td>AF</td>\n",
       "      <td>Kabul</td>\n",
       "      <td>EUR</td>\n",
       "      <td>S����</td>\n",
       "      <td>English</td>\n",
       "      <td>Africa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>freq</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>23</td>\n",
       "      <td>2</td>\n",
       "      <td>31</td>\n",
       "      <td>53</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       abbreviation capital/majorcity currency-code largestcity  \\\n",
       "count           188               192           180         189   \n",
       "unique          188               192           133         188   \n",
       "top              AF             Kabul           EUR       S����   \n",
       "freq              1                 1            23           2   \n",
       "\n",
       "       officiallanguage continent  \n",
       "count               190       180  \n",
       "unique               76         6  \n",
       "top             English    Africa  \n",
       "freq                 31        53  "
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "stats_descriptivas"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "No tiene sentido substituir estos valores por la media o por la moda, por lo que creamos la categotía \"Unknow\" para clasificar todos estos valores nulos:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "columnas_unknow = ['abbreviation', 'capital/majorcity', 'currency-code', 'largestcity',\n",
    "       'officiallanguage', 'continent', 'minimumwage']\n",
    "\n",
    "# Nota: Hemos añadido \"minimum wage\" a la lista, ya que también se tenía que cambiar los nulos a categoría \"Unknown\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Después del reemplazo usando 'fillna' quedan los siguientes nulos\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "abbreviation         0\n",
       "capital/majorcity    0\n",
       "currency-code        0\n",
       "largestcity          0\n",
       "officiallanguage     0\n",
       "continent            0\n",
       "minimumwage          0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "for columna in columnas_unknow:\n",
    "    \n",
    "    # reemplazamos los nulos por el valor Unknown para cada una de las columnas de la lista\n",
    "    df_world[columna] = df_world[columna].fillna(\"Unknown\")\n",
    "    \n",
    "# comprobamos si quedan nulos en las columnas categóricas. \n",
    "print(\"Después del reemplazo usando 'fillna' quedan los siguientes nulos\")\n",
    "df_world[columnas_unknow].isnull().sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "______\n",
    "\n",
    "## Variables numéricas"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Las columnas numéricas que tienen nulos son : \n",
      " \n",
      "Index(['agriculturalland', 'landarea', 'armedforcessize', 'birthrate',\n",
      "       'callingcode', 'co2-emissions', 'cpi', 'cpichange', 'fertilityrate',\n",
      "       'forestedarea', 'gasolineprice', 'gdp',\n",
      "       'grossprimaryeducationenrollment', 'grosstertiaryeducationenrollment',\n",
      "       'infantmortality', 'lifeexpectancy', 'maternalmortalityratio',\n",
      "       'outofpockethealthexpenditure', 'physiciansperthousand', 'population',\n",
      "       'populationlaborforceparticipation', 'taxrevenue', 'totaltaxrate',\n",
      "       'unemploymentrate', 'urban_population', 'longitud', 'latitud'],\n",
      "      dtype='object')\n"
     ]
    }
   ],
   "source": [
    "# Obtenemos la lista de columnas numericas que tienen nulos\n",
    "nulos_esta_num = df_world[df_world.columns[df_world.isnull().any()]].select_dtypes(include = np.number).columns\n",
    "print(\"Las columnas numéricas que tienen nulos son : \\n \")\n",
    "print(nulos_esta_num)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "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>%nulos</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>agriculturalland</th>\n",
       "      <td>3.589744</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>landarea</th>\n",
       "      <td>0.512821</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>armedforcessize</th>\n",
       "      <td>12.307692</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>birthrate</th>\n",
       "      <td>3.076923</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>callingcode</th>\n",
       "      <td>0.512821</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>co2-emissions</th>\n",
       "      <td>3.589744</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>cpi</th>\n",
       "      <td>8.717949</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>cpichange</th>\n",
       "      <td>8.205128</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>fertilityrate</th>\n",
       "      <td>3.589744</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>forestedarea</th>\n",
       "      <td>3.589744</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>gasolineprice</th>\n",
       "      <td>10.256410</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>gdp</th>\n",
       "      <td>1.025641</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>grossprimaryeducationenrollment</th>\n",
       "      <td>3.589744</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>grosstertiaryeducationenrollment</th>\n",
       "      <td>6.153846</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>infantmortality</th>\n",
       "      <td>3.076923</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>lifeexpectancy</th>\n",
       "      <td>4.102564</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>maternalmortalityratio</th>\n",
       "      <td>7.179487</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>outofpockethealthexpenditure</th>\n",
       "      <td>3.589744</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>physiciansperthousand</th>\n",
       "      <td>3.589744</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>population</th>\n",
       "      <td>0.512821</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>populationlaborforceparticipation</th>\n",
       "      <td>9.743590</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>taxrevenue</th>\n",
       "      <td>13.333333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>totaltaxrate</th>\n",
       "      <td>6.153846</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>unemploymentrate</th>\n",
       "      <td>9.743590</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>urban_population</th>\n",
       "      <td>2.564103</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>longitud</th>\n",
       "      <td>0.512821</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>latitud</th>\n",
       "      <td>0.512821</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                      %nulos\n",
       "agriculturalland                    3.589744\n",
       "landarea                            0.512821\n",
       "armedforcessize                    12.307692\n",
       "birthrate                           3.076923\n",
       "callingcode                         0.512821\n",
       "co2-emissions                       3.589744\n",
       "cpi                                 8.717949\n",
       "cpichange                           8.205128\n",
       "fertilityrate                       3.589744\n",
       "forestedarea                        3.589744\n",
       "gasolineprice                      10.256410\n",
       "gdp                                 1.025641\n",
       "grossprimaryeducationenrollment     3.589744\n",
       "grosstertiaryeducationenrollment    6.153846\n",
       "infantmortality                     3.076923\n",
       "lifeexpectancy                      4.102564\n",
       "maternalmortalityratio              7.179487\n",
       "outofpockethealthexpenditure        3.589744\n",
       "physiciansperthousand               3.589744\n",
       "population                          0.512821\n",
       "populationlaborforceparticipation   9.743590\n",
       "taxrevenue                         13.333333\n",
       "totaltaxrate                        6.153846\n",
       "unemploymentrate                    9.743590\n",
       "urban_population                    2.564103\n",
       "longitud                            0.512821\n",
       "latitud                             0.512821"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Calcularmos el porcentaje de nulos que tenemos en cada una de las columnas numéricas\n",
    "# Pasamos los datos a Dataframe y llamamos a la columa con los datos '%nulos'\n",
    "df_porcentaje_nulos_numericos = pd.DataFrame(df_world[nulos_esta_num].isnull().sum() / df_world.shape[0] * 100, columns=['%nulos'])\n",
    "df_porcentaje_nulos_numericos"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Criterio para gestionar los nulos de las variables numéricas:**\n",
    "\n",
    "1 - Las columnas con porcentajes inferiores o igual al 5% se pueden imputar por la media.\n",
    "\n",
    "2- Aquellas columnas que tienen valores entre el 5-10% podemos optar por reemplazar por la media o mediana. Haremos una comparativa para valorar si se imputa con KNN Imputer o co Iterative Imputer\n",
    "\n",
    "3- Hay tres columnas con valores superiores o igual al 10%: \n",
    "- armedforcessize\n",
    "- gasolineprice\n",
    "- taxrevenue"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "______________\n",
    "1 -  Columnas con porcentajes inferiores o igual al 5%:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['agriculturalland', 'landarea', 'birthrate', 'callingcode',\n",
       "       'co2-emissions', 'fertilityrate', 'forestedarea', 'gdp',\n",
       "       'grossprimaryeducationenrollment', 'infantmortality', 'lifeexpectancy',\n",
       "       'outofpockethealthexpenditure', 'physiciansperthousand', 'population',\n",
       "       'urban_population', 'longitud', 'latitud'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "condicion_1 = df_porcentaje_nulos_numericos.loc[(df_porcentaje_nulos_numericos['%nulos']<= 5) & (df_porcentaje_nulos_numericos['%nulos']> 0)]\n",
    "condicion_1.index"
   ]
  },
  {
   "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></th>\n",
       "      <th>%nulos</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>agriculturalland</th>\n",
       "      <td>3.589744</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>landarea</th>\n",
       "      <td>0.512821</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>birthrate</th>\n",
       "      <td>3.076923</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>callingcode</th>\n",
       "      <td>0.512821</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>co2-emissions</th>\n",
       "      <td>3.589744</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>fertilityrate</th>\n",
       "      <td>3.589744</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>forestedarea</th>\n",
       "      <td>3.589744</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>gdp</th>\n",
       "      <td>1.025641</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>grossprimaryeducationenrollment</th>\n",
       "      <td>3.589744</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>infantmortality</th>\n",
       "      <td>3.076923</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>lifeexpectancy</th>\n",
       "      <td>4.102564</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>outofpockethealthexpenditure</th>\n",
       "      <td>3.589744</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>physiciansperthousand</th>\n",
       "      <td>3.589744</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>population</th>\n",
       "      <td>0.512821</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>urban_population</th>\n",
       "      <td>2.564103</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>longitud</th>\n",
       "      <td>0.512821</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>latitud</th>\n",
       "      <td>0.512821</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                   %nulos\n",
       "agriculturalland                 3.589744\n",
       "landarea                         0.512821\n",
       "birthrate                        3.076923\n",
       "callingcode                      0.512821\n",
       "co2-emissions                    3.589744\n",
       "fertilityrate                    3.589744\n",
       "forestedarea                     3.589744\n",
       "gdp                              1.025641\n",
       "grossprimaryeducationenrollment  3.589744\n",
       "infantmortality                  3.076923\n",
       "lifeexpectancy                   4.102564\n",
       "outofpockethealthexpenditure     3.589744\n",
       "physiciansperthousand            3.589744\n",
       "population                       0.512821\n",
       "urban_population                 2.564103\n",
       "longitud                         0.512821\n",
       "latitud                          0.512821"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "condicion_1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "lista_condicion_1 = ['agriculturalland', 'landarea', 'birthrate', 'callingcode',\n",
    "       'co2-emissions', 'fertilityrate', 'forestedarea', 'gdp',\n",
    "       'grossprimaryeducationenrollment', 'infantmortality', 'lifeexpectancy',\n",
    "       'outofpockethealthexpenditure', 'physiciansperthousand', 'population',\n",
    "       'urban_population', 'longitud', 'latitud']"
   ]
  },
  {
   "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>agriculturalland</th>\n",
       "      <th>landarea</th>\n",
       "      <th>birthrate</th>\n",
       "      <th>callingcode</th>\n",
       "      <th>co2-emissions</th>\n",
       "      <th>fertilityrate</th>\n",
       "      <th>forestedarea</th>\n",
       "      <th>gdp</th>\n",
       "      <th>grossprimaryeducationenrollment</th>\n",
       "      <th>infantmortality</th>\n",
       "      <th>lifeexpectancy</th>\n",
       "      <th>outofpockethealthexpenditure</th>\n",
       "      <th>physiciansperthousand</th>\n",
       "      <th>population</th>\n",
       "      <th>urban_population</th>\n",
       "      <th>longitud</th>\n",
       "      <th>latitud</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>188.000000</td>\n",
       "      <td>1.940000e+02</td>\n",
       "      <td>189.000000</td>\n",
       "      <td>194.000000</td>\n",
       "      <td>1.880000e+02</td>\n",
       "      <td>188.000000</td>\n",
       "      <td>188.000000</td>\n",
       "      <td>1.930000e+02</td>\n",
       "      <td>188.000000</td>\n",
       "      <td>189.000000</td>\n",
       "      <td>187.000000</td>\n",
       "      <td>188.000000</td>\n",
       "      <td>188.000000</td>\n",
       "      <td>1.940000e+02</td>\n",
       "      <td>1.900000e+02</td>\n",
       "      <td>194.000000</td>\n",
       "      <td>194.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>39.117553</td>\n",
       "      <td>6.896244e+05</td>\n",
       "      <td>20.214974</td>\n",
       "      <td>360.546392</td>\n",
       "      <td>1.777992e+05</td>\n",
       "      <td>2.698138</td>\n",
       "      <td>32.015426</td>\n",
       "      <td>4.772959e+11</td>\n",
       "      <td>102.470213</td>\n",
       "      <td>21.332804</td>\n",
       "      <td>72.279679</td>\n",
       "      <td>32.637234</td>\n",
       "      <td>1.839840</td>\n",
       "      <td>3.938116e+07</td>\n",
       "      <td>2.230454e+07</td>\n",
       "      <td>19.092351</td>\n",
       "      <td>20.232434</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>21.783052</td>\n",
       "      <td>1.921609e+06</td>\n",
       "      <td>9.945774</td>\n",
       "      <td>323.236419</td>\n",
       "      <td>8.387903e+05</td>\n",
       "      <td>1.282267</td>\n",
       "      <td>23.791331</td>\n",
       "      <td>2.172173e+12</td>\n",
       "      <td>13.151607</td>\n",
       "      <td>19.548058</td>\n",
       "      <td>7.483661</td>\n",
       "      <td>19.172697</td>\n",
       "      <td>1.684261</td>\n",
       "      <td>1.450924e+08</td>\n",
       "      <td>7.543050e+07</td>\n",
       "      <td>23.961779</td>\n",
       "      <td>66.716110</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.600000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>5.900000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.100000e+01</td>\n",
       "      <td>0.980000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>4.727146e+07</td>\n",
       "      <td>23.400000</td>\n",
       "      <td>1.400000</td>\n",
       "      <td>52.800000</td>\n",
       "      <td>0.200000</td>\n",
       "      <td>0.010000</td>\n",
       "      <td>8.360000e+02</td>\n",
       "      <td>5.464000e+03</td>\n",
       "      <td>-40.900557</td>\n",
       "      <td>-175.198242</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>21.700000</td>\n",
       "      <td>2.382825e+04</td>\n",
       "      <td>11.300000</td>\n",
       "      <td>82.500000</td>\n",
       "      <td>2.304250e+03</td>\n",
       "      <td>1.705000</td>\n",
       "      <td>11.000000</td>\n",
       "      <td>8.454620e+09</td>\n",
       "      <td>98.950000</td>\n",
       "      <td>6.000000</td>\n",
       "      <td>67.000000</td>\n",
       "      <td>17.450000</td>\n",
       "      <td>0.332500</td>\n",
       "      <td>1.962678e+06</td>\n",
       "      <td>1.152961e+06</td>\n",
       "      <td>4.544175</td>\n",
       "      <td>-7.941496</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>39.600000</td>\n",
       "      <td>1.195110e+05</td>\n",
       "      <td>17.950000</td>\n",
       "      <td>255.500000</td>\n",
       "      <td>1.230300e+04</td>\n",
       "      <td>2.245000</td>\n",
       "      <td>32.000000</td>\n",
       "      <td>3.438723e+10</td>\n",
       "      <td>102.550000</td>\n",
       "      <td>14.000000</td>\n",
       "      <td>73.200000</td>\n",
       "      <td>30.700000</td>\n",
       "      <td>1.460000</td>\n",
       "      <td>8.826588e+06</td>\n",
       "      <td>4.678104e+06</td>\n",
       "      <td>17.273849</td>\n",
       "      <td>20.972652</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>55.375000</td>\n",
       "      <td>5.242560e+05</td>\n",
       "      <td>28.750000</td>\n",
       "      <td>506.750000</td>\n",
       "      <td>6.388425e+04</td>\n",
       "      <td>3.597500</td>\n",
       "      <td>48.175000</td>\n",
       "      <td>2.340940e+11</td>\n",
       "      <td>108.025000</td>\n",
       "      <td>32.700000</td>\n",
       "      <td>77.500000</td>\n",
       "      <td>43.975000</td>\n",
       "      <td>2.935000</td>\n",
       "      <td>2.858549e+07</td>\n",
       "      <td>1.490324e+07</td>\n",
       "      <td>40.124603</td>\n",
       "      <td>48.281523</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>82.600000</td>\n",
       "      <td>1.709824e+07</td>\n",
       "      <td>46.080000</td>\n",
       "      <td>1876.000000</td>\n",
       "      <td>9.893038e+06</td>\n",
       "      <td>6.910000</td>\n",
       "      <td>98.300000</td>\n",
       "      <td>2.142770e+13</td>\n",
       "      <td>142.500000</td>\n",
       "      <td>84.500000</td>\n",
       "      <td>85.400000</td>\n",
       "      <td>81.600000</td>\n",
       "      <td>8.420000</td>\n",
       "      <td>1.397715e+09</td>\n",
       "      <td>8.429340e+08</td>\n",
       "      <td>64.963051</td>\n",
       "      <td>178.065032</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       agriculturalland      landarea   birthrate  callingcode  co2-emissions  \\\n",
       "count        188.000000  1.940000e+02  189.000000   194.000000   1.880000e+02   \n",
       "mean          39.117553  6.896244e+05   20.214974   360.546392   1.777992e+05   \n",
       "std           21.783052  1.921609e+06    9.945774   323.236419   8.387903e+05   \n",
       "min            0.600000  0.000000e+00    5.900000     1.000000   1.100000e+01   \n",
       "25%           21.700000  2.382825e+04   11.300000    82.500000   2.304250e+03   \n",
       "50%           39.600000  1.195110e+05   17.950000   255.500000   1.230300e+04   \n",
       "75%           55.375000  5.242560e+05   28.750000   506.750000   6.388425e+04   \n",
       "max           82.600000  1.709824e+07   46.080000  1876.000000   9.893038e+06   \n",
       "\n",
       "       fertilityrate  forestedarea           gdp  \\\n",
       "count     188.000000    188.000000  1.930000e+02   \n",
       "mean        2.698138     32.015426  4.772959e+11   \n",
       "std         1.282267     23.791331  2.172173e+12   \n",
       "min         0.980000      0.000000  4.727146e+07   \n",
       "25%         1.705000     11.000000  8.454620e+09   \n",
       "50%         2.245000     32.000000  3.438723e+10   \n",
       "75%         3.597500     48.175000  2.340940e+11   \n",
       "max         6.910000     98.300000  2.142770e+13   \n",
       "\n",
       "       grossprimaryeducationenrollment  infantmortality  lifeexpectancy  \\\n",
       "count                       188.000000       189.000000      187.000000   \n",
       "mean                        102.470213        21.332804       72.279679   \n",
       "std                          13.151607        19.548058        7.483661   \n",
       "min                          23.400000         1.400000       52.800000   \n",
       "25%                          98.950000         6.000000       67.000000   \n",
       "50%                         102.550000        14.000000       73.200000   \n",
       "75%                         108.025000        32.700000       77.500000   \n",
       "max                         142.500000        84.500000       85.400000   \n",
       "\n",
       "       outofpockethealthexpenditure  physiciansperthousand    population  \\\n",
       "count                    188.000000             188.000000  1.940000e+02   \n",
       "mean                      32.637234               1.839840  3.938116e+07   \n",
       "std                       19.172697               1.684261  1.450924e+08   \n",
       "min                        0.200000               0.010000  8.360000e+02   \n",
       "25%                       17.450000               0.332500  1.962678e+06   \n",
       "50%                       30.700000               1.460000  8.826588e+06   \n",
       "75%                       43.975000               2.935000  2.858549e+07   \n",
       "max                       81.600000               8.420000  1.397715e+09   \n",
       "\n",
       "       urban_population    longitud     latitud  \n",
       "count      1.900000e+02  194.000000  194.000000  \n",
       "mean       2.230454e+07   19.092351   20.232434  \n",
       "std        7.543050e+07   23.961779   66.716110  \n",
       "min        5.464000e+03  -40.900557 -175.198242  \n",
       "25%        1.152961e+06    4.544175   -7.941496  \n",
       "50%        4.678104e+06   17.273849   20.972652  \n",
       "75%        1.490324e+07   40.124603   48.281523  \n",
       "max        8.429340e+08   64.963051  178.065032  "
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# para decidir si usamos la media o la mediana lo que haremos es hacer un describe de la columna para saber cuales son estos valores\n",
    "\n",
    "df_world[lista_condicion_1].describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Rellenamos por la media:\n",
    "\n",
    "df_world[lista_condicion_1] = df_world[lista_condicion_1].fillna(df_world[lista_condicion_1].mean())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Después del 'fillna' tenemos agriculturalland                   0\n",
      "landarea                           0\n",
      "birthrate                          0\n",
      "callingcode                        0\n",
      "co2-emissions                      0\n",
      "fertilityrate                      0\n",
      "forestedarea                       0\n",
      "gdp                                0\n",
      "grossprimaryeducationenrollment    0\n",
      "infantmortality                    0\n",
      "lifeexpectancy                     0\n",
      "outofpockethealthexpenditure       0\n",
      "physiciansperthousand              0\n",
      "population                         0\n",
      "urban_population                   0\n",
      "longitud                           0\n",
      "latitud                            0\n",
      "dtype: int64 nulos\n"
     ]
    }
   ],
   "source": [
    "# comprobamos los nulos para la columna\n",
    "print(f\"Después del 'fillna' tenemos {df_world[lista_condicion_1].isnull().sum()} nulos\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "______\n",
    "\n",
    "2- Gestionamos las columnas que estan entre el 5 y el 10% de nulos.\n",
    "Haremos una comparativa para valorar si se imputa con KNN Imputer o co Iterative Imputer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['cpi', 'cpichange', 'grosstertiaryeducationenrollment',\n",
       "       'maternalmortalityratio', 'populationlaborforceparticipation',\n",
       "       'totaltaxrate', 'unemploymentrate'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "condicion_2 = df_porcentaje_nulos_numericos.loc[(df_porcentaje_nulos_numericos['%nulos']> 5) & (df_porcentaje_nulos_numericos['%nulos']< 10)]\n",
    "condicion_2.index"
   ]
  },
  {
   "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>country</th>\n",
       "      <th>density</th>\n",
       "      <th>abbreviation</th>\n",
       "      <th>agriculturalland</th>\n",
       "      <th>landarea</th>\n",
       "      <th>armedforcessize</th>\n",
       "      <th>birthrate</th>\n",
       "      <th>callingcode</th>\n",
       "      <th>capital/majorcity</th>\n",
       "      <th>co2-emissions</th>\n",
       "      <th>cpi</th>\n",
       "      <th>cpichange</th>\n",
       "      <th>currency-code</th>\n",
       "      <th>fertilityrate</th>\n",
       "      <th>forestedarea</th>\n",
       "      <th>gasolineprice</th>\n",
       "      <th>gdp</th>\n",
       "      <th>grossprimaryeducationenrollment</th>\n",
       "      <th>grosstertiaryeducationenrollment</th>\n",
       "      <th>infantmortality</th>\n",
       "      <th>largestcity</th>\n",
       "      <th>lifeexpectancy</th>\n",
       "      <th>maternalmortalityratio</th>\n",
       "      <th>minimumwage</th>\n",
       "      <th>officiallanguage</th>\n",
       "      <th>outofpockethealthexpenditure</th>\n",
       "      <th>physiciansperthousand</th>\n",
       "      <th>population</th>\n",
       "      <th>populationlaborforceparticipation</th>\n",
       "      <th>taxrevenue</th>\n",
       "      <th>totaltaxrate</th>\n",
       "      <th>unemploymentrate</th>\n",
       "      <th>urban_population</th>\n",
       "      <th>longitud</th>\n",
       "      <th>latitud</th>\n",
       "      <th>continent</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Afghanistan</td>\n",
       "      <td>60</td>\n",
       "      <td>AF</td>\n",
       "      <td>58.1</td>\n",
       "      <td>652230.0</td>\n",
       "      <td>323000.0</td>\n",
       "      <td>32.49</td>\n",
       "      <td>93.0</td>\n",
       "      <td>Kabul</td>\n",
       "      <td>8672.0</td>\n",
       "      <td>149.9</td>\n",
       "      <td>2.3</td>\n",
       "      <td>AFN</td>\n",
       "      <td>4.47</td>\n",
       "      <td>2.1</td>\n",
       "      <td>0.7</td>\n",
       "      <td>1.910135e+10</td>\n",
       "      <td>104.0</td>\n",
       "      <td>9.7</td>\n",
       "      <td>47.9</td>\n",
       "      <td>Kabul</td>\n",
       "      <td>64.5</td>\n",
       "      <td>638.0</td>\n",
       "      <td>0.43</td>\n",
       "      <td>Pashto</td>\n",
       "      <td>78.4</td>\n",
       "      <td>0.28</td>\n",
       "      <td>38041754.0</td>\n",
       "      <td>48.9</td>\n",
       "      <td>9.3</td>\n",
       "      <td>71.4</td>\n",
       "      <td>11.12</td>\n",
       "      <td>9797273.0</td>\n",
       "      <td>33.93911</td>\n",
       "      <td>67.709953</td>\n",
       "      <td>Asia</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       country density abbreviation  agriculturalland  landarea  \\\n",
       "0  Afghanistan      60           AF              58.1  652230.0   \n",
       "\n",
       "   armedforcessize  birthrate  callingcode capital/majorcity  co2-emissions  \\\n",
       "0         323000.0      32.49         93.0             Kabul         8672.0   \n",
       "\n",
       "     cpi  cpichange currency-code  fertilityrate  forestedarea  gasolineprice  \\\n",
       "0  149.9        2.3           AFN           4.47           2.1            0.7   \n",
       "\n",
       "            gdp  grossprimaryeducationenrollment  \\\n",
       "0  1.910135e+10                            104.0   \n",
       "\n",
       "   grosstertiaryeducationenrollment  infantmortality largestcity  \\\n",
       "0                               9.7             47.9       Kabul   \n",
       "\n",
       "   lifeexpectancy  maternalmortalityratio minimumwage officiallanguage  \\\n",
       "0            64.5                   638.0        0.43           Pashto   \n",
       "\n",
       "   outofpockethealthexpenditure  physiciansperthousand  population  \\\n",
       "0                          78.4                   0.28  38041754.0   \n",
       "\n",
       "   populationlaborforceparticipation  taxrevenue  totaltaxrate  \\\n",
       "0                               48.9         9.3          71.4   \n",
       "\n",
       "   unemploymentrate  urban_population  longitud    latitud continent  \n",
       "0             11.12         9797273.0  33.93911  67.709953      Asia  "
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# como vamos a crear columnas nuevas, vamos a hacer una copia del DataFrame usando el método 'copy()'\n",
    "df_world_copia = df_world.copy()\n",
    "df_world_copia .head(1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "IterativeImputer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/lluerna/opt/anaconda3/lib/python3.9/site-packages/sklearn/utils/validation.py:571: FutureWarning: is_sparse is deprecated and will be removed in a future version. Check `isinstance(dtype, pd.SparseDtype)` instead.\n",
      "  array.dtypes.apply(is_sparse).any()):\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "array([[149.9 ,   2.3 ,   9.7 , ...,  48.9 ,  71.4 ,  11.12],\n",
       "       [119.05,   1.4 ,  55.  , ...,  55.7 ,  36.6 ,  12.33],\n",
       "       [151.36,   2.  ,  51.4 , ...,  41.2 ,  66.1 ,  11.7 ],\n",
       "       ...,\n",
       "       [157.58,   8.1 ,  10.2 , ...,  38.  ,  26.6 ,  12.91],\n",
       "       [212.31,   9.2 ,   4.1 , ...,  74.6 ,  15.6 ,  11.43],\n",
       "       [105.51,   0.9 ,  10.  , ...,  83.1 ,  31.6 ,   4.95]])"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# instanciamos las clases\n",
    "imputer_iterative = IterativeImputer(max_iter = 20, random_state = 42)\n",
    "\n",
    "# ajustamos y tranformamos los datos\n",
    "imputer_iterative_imputado = imputer_iterative.fit_transform(df_world_copia[['cpi', 'cpichange', 'grosstertiaryeducationenrollment',\n",
    "       'maternalmortalityratio', 'populationlaborforceparticipation',\n",
    "       'totaltaxrate', 'unemploymentrate']])\n",
    "\n",
    "# comprobamos que es lo que nos devuelve, que en este caso es un array también\n",
    "imputer_iterative_imputado"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Después del 'Iterative' tenemos: \n",
      "cpi_iterative                                  0\n",
      "cpichange_iterative                            0\n",
      "grosstertiaryeducationenrollment_iterative     0\n",
      "maternalmortalityratio_iterative               0\n",
      "populationlaborforceparticipation_iterative    0\n",
      "totaltaxrate_iterative                         0\n",
      "unemploymentrate_iterative                     0\n",
      "dtype: int64 nulos\n"
     ]
    }
   ],
   "source": [
    "# en este caso como vamos a aplicar los dos métodos sobre estas columnas vamos a crear dos columnas nuevas\n",
    "df_world_copia[['cpi_iterative', 'cpichange_iterative', 'grosstertiaryeducationenrollment_iterative', 'maternalmortalityratio_iterative', 'populationlaborforceparticipation_iterative', 'totaltaxrate_iterative', 'unemploymentrate_iterative']] = imputer_iterative_imputado\n",
    "\n",
    "# comprobamos los nulos\n",
    "print(f\"Después del 'Iterative' tenemos: \\n{df_world_copia[['cpi_iterative', 'cpichange_iterative', 'grosstertiaryeducationenrollment_iterative', 'maternalmortalityratio_iterative', 'populationlaborforceparticipation_iterative', 'totaltaxrate_iterative', 'unemploymentrate_iterative']].isnull().sum()} nulos\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "KNNImputer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/lluerna/opt/anaconda3/lib/python3.9/site-packages/sklearn/utils/validation.py:571: FutureWarning: is_sparse is deprecated and will be removed in a future version. Check `isinstance(dtype, pd.SparseDtype)` instead.\n",
      "  array.dtypes.apply(is_sparse).any()):\n",
      "/Users/lluerna/opt/anaconda3/lib/python3.9/site-packages/sklearn/utils/validation.py:571: FutureWarning: is_sparse is deprecated and will be removed in a future version. Check `isinstance(dtype, pd.SparseDtype)` instead.\n",
      "  array.dtypes.apply(is_sparse).any()):\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "array([[149.9 ,   2.3 ,   9.7 , ...,  48.9 ,  71.4 ,  11.12],\n",
       "       [119.05,   1.4 ,  55.  , ...,  55.7 ,  36.6 ,  12.33],\n",
       "       [151.36,   2.  ,  51.4 , ...,  41.2 ,  66.1 ,  11.7 ],\n",
       "       ...,\n",
       "       [157.58,   8.1 ,  10.2 , ...,  38.  ,  26.6 ,  12.91],\n",
       "       [212.31,   9.2 ,   4.1 , ...,  74.6 ,  15.6 ,  11.43],\n",
       "       [105.51,   0.9 ,  10.  , ...,  83.1 ,  31.6 ,   4.95]])"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Hacemos lo mismo con el KNNImputer. \n",
    "\n",
    "# instanciamos la clase del KNNImputer\n",
    "imputer_knn = KNNImputer(n_neighbors = 5)\n",
    "\n",
    "# ajustamos y transformamos los datos\n",
    "imputer_knn_imputado = imputer_knn.fit_transform(df_world_copia[['cpi', 'cpichange', 'grosstertiaryeducationenrollment', 'maternalmortalityratio', 'populationlaborforceparticipation','totaltaxrate', 'unemploymentrate']])\n",
    "\n",
    "# comprobamos que es lo que nos devuelve, que sigue siendo un array\n",
    "imputer_knn_imputado"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Después del 'KNN' tenemos: \n",
      "cpi_knn                                  0\n",
      "cpichange_knn                            0\n",
      "grosstertiaryeducationenrollment_knn     0\n",
      "maternalmortalityratio_knn               0\n",
      "populationlaborforceparticipation_knn    0\n",
      "totaltaxrate_knn                         0\n",
      "unemploymentrate_knn                     0\n",
      "dtype: int64 nulos\n"
     ]
    }
   ],
   "source": [
    "# por último nos queda añadir ese array al DataFrame como hemos hecho hasta ahora\n",
    "df_world_copia[['cpi_knn', 'cpichange_knn', 'grosstertiaryeducationenrollment_knn', 'maternalmortalityratio_knn', 'populationlaborforceparticipation_knn','totaltaxrate_knn', 'unemploymentrate_knn']] = imputer_knn_imputado\n",
    "\n",
    "\n",
    "# comprobamos los nulos\n",
    "print(f\"Después del 'KNN' tenemos: \\n{df_world_copia[['cpi_knn', 'cpichange_knn', 'grosstertiaryeducationenrollment_knn', 'maternalmortalityratio_knn', 'populationlaborforceparticipation_knn','totaltaxrate_knn', 'unemploymentrate_knn']].isnull().sum()} nulos\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Comparamos los datos con ambos metodos para decidir con cual nos quedamos. En función del método de imputación que más se asemeja al dato original. \n",
    "\n",
    "Como hay muchas columnas, lo miramos en dos partes:\n",
    "\n",
    "1:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "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>cpi</th>\n",
       "      <th>cpi_knn</th>\n",
       "      <th>cpi_iterative</th>\n",
       "      <th>cpichange</th>\n",
       "      <th>cpichange_knn</th>\n",
       "      <th>cpichange_iterative</th>\n",
       "      <th>grosstertiaryeducationenrollment</th>\n",
       "      <th>grosstertiaryeducationenrollment_knn</th>\n",
       "      <th>grosstertiaryeducationenrollment_iterative</th>\n",
       "      <th>maternalmortalityratio</th>\n",
       "      <th>maternalmortalityratio_knn</th>\n",
       "      <th>maternalmortalityratio_iterative</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>178.000000</td>\n",
       "      <td>195.000000</td>\n",
       "      <td>195.000000</td>\n",
       "      <td>179.000000</td>\n",
       "      <td>195.000000</td>\n",
       "      <td>195.000000</td>\n",
       "      <td>183.000000</td>\n",
       "      <td>195.000000</td>\n",
       "      <td>195.000000</td>\n",
       "      <td>181.000000</td>\n",
       "      <td>195.000000</td>\n",
       "      <td>195.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>190.460955</td>\n",
       "      <td>188.044542</td>\n",
       "      <td>189.733384</td>\n",
       "      <td>6.722346</td>\n",
       "      <td>6.635378</td>\n",
       "      <td>6.684134</td>\n",
       "      <td>37.963388</td>\n",
       "      <td>37.912344</td>\n",
       "      <td>37.768805</td>\n",
       "      <td>160.392265</td>\n",
       "      <td>157.789426</td>\n",
       "      <td>160.021942</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>397.947380</td>\n",
       "      <td>380.285979</td>\n",
       "      <td>381.626934</td>\n",
       "      <td>24.450414</td>\n",
       "      <td>23.435360</td>\n",
       "      <td>23.483419</td>\n",
       "      <td>29.261022</td>\n",
       "      <td>28.357205</td>\n",
       "      <td>28.530212</td>\n",
       "      <td>233.502024</td>\n",
       "      <td>226.006843</td>\n",
       "      <td>225.688610</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>99.030000</td>\n",
       "      <td>99.030000</td>\n",
       "      <td>31.645208</td>\n",
       "      <td>-4.300000</td>\n",
       "      <td>-4.300000</td>\n",
       "      <td>-4.300000</td>\n",
       "      <td>0.800000</td>\n",
       "      <td>0.800000</td>\n",
       "      <td>-7.409868</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>-45.950535</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>113.885000</td>\n",
       "      <td>114.415000</td>\n",
       "      <td>113.960000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>12.150000</td>\n",
       "      <td>12.650000</td>\n",
       "      <td>12.500000</td>\n",
       "      <td>13.000000</td>\n",
       "      <td>15.500000</td>\n",
       "      <td>15.500000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>125.340000</td>\n",
       "      <td>129.000000</td>\n",
       "      <td>127.630000</td>\n",
       "      <td>2.300000</td>\n",
       "      <td>2.500000</td>\n",
       "      <td>2.500000</td>\n",
       "      <td>31.200000</td>\n",
       "      <td>34.400000</td>\n",
       "      <td>34.600000</td>\n",
       "      <td>53.000000</td>\n",
       "      <td>60.000000</td>\n",
       "      <td>65.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>157.265000</td>\n",
       "      <td>162.605000</td>\n",
       "      <td>166.690000</td>\n",
       "      <td>4.250000</td>\n",
       "      <td>5.200000</td>\n",
       "      <td>5.500000</td>\n",
       "      <td>63.150000</td>\n",
       "      <td>61.100000</td>\n",
       "      <td>61.100000</td>\n",
       "      <td>186.000000</td>\n",
       "      <td>180.000000</td>\n",
       "      <td>184.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>4583.710000</td>\n",
       "      <td>4583.710000</td>\n",
       "      <td>4583.710000</td>\n",
       "      <td>254.900000</td>\n",
       "      <td>254.900000</td>\n",
       "      <td>254.900000</td>\n",
       "      <td>136.600000</td>\n",
       "      <td>136.600000</td>\n",
       "      <td>136.600000</td>\n",
       "      <td>1150.000000</td>\n",
       "      <td>1150.000000</td>\n",
       "      <td>1150.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               cpi      cpi_knn  cpi_iterative   cpichange  cpichange_knn  \\\n",
       "count   178.000000   195.000000     195.000000  179.000000     195.000000   \n",
       "mean    190.460955   188.044542     189.733384    6.722346       6.635378   \n",
       "std     397.947380   380.285979     381.626934   24.450414      23.435360   \n",
       "min      99.030000    99.030000      31.645208   -4.300000      -4.300000   \n",
       "25%     113.885000   114.415000     113.960000    1.000000       1.000000   \n",
       "50%     125.340000   129.000000     127.630000    2.300000       2.500000   \n",
       "75%     157.265000   162.605000     166.690000    4.250000       5.200000   \n",
       "max    4583.710000  4583.710000    4583.710000  254.900000     254.900000   \n",
       "\n",
       "       cpichange_iterative  grosstertiaryeducationenrollment  \\\n",
       "count           195.000000                        183.000000   \n",
       "mean              6.684134                         37.963388   \n",
       "std              23.483419                         29.261022   \n",
       "min              -4.300000                          0.800000   \n",
       "25%               1.000000                         12.150000   \n",
       "50%               2.500000                         31.200000   \n",
       "75%               5.500000                         63.150000   \n",
       "max             254.900000                        136.600000   \n",
       "\n",
       "       grosstertiaryeducationenrollment_knn  \\\n",
       "count                            195.000000   \n",
       "mean                              37.912344   \n",
       "std                               28.357205   \n",
       "min                                0.800000   \n",
       "25%                               12.650000   \n",
       "50%                               34.400000   \n",
       "75%                               61.100000   \n",
       "max                              136.600000   \n",
       "\n",
       "       grosstertiaryeducationenrollment_iterative  maternalmortalityratio  \\\n",
       "count                                  195.000000              181.000000   \n",
       "mean                                    37.768805              160.392265   \n",
       "std                                     28.530212              233.502024   \n",
       "min                                     -7.409868                2.000000   \n",
       "25%                                     12.500000               13.000000   \n",
       "50%                                     34.600000               53.000000   \n",
       "75%                                     61.100000              186.000000   \n",
       "max                                    136.600000             1150.000000   \n",
       "\n",
       "       maternalmortalityratio_knn  maternalmortalityratio_iterative  \n",
       "count                  195.000000                        195.000000  \n",
       "mean                   157.789426                        160.021942  \n",
       "std                    226.006843                        225.688610  \n",
       "min                      2.000000                        -45.950535  \n",
       "25%                     15.500000                         15.500000  \n",
       "50%                     60.000000                         65.000000  \n",
       "75%                    180.000000                        184.000000  \n",
       "max                   1150.000000                       1150.000000  "
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_world_copia.describe()[['cpi','cpi_knn', 'cpi_iterative', 'cpichange', 'cpichange_knn','cpichange_iterative', 'grosstertiaryeducationenrollment','grosstertiaryeducationenrollment_knn','grosstertiaryeducationenrollment_iterative', 'maternalmortalityratio', 'maternalmortalityratio_knn', 'maternalmortalityratio_iterative']]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- cpi: Imputaremos por iterative\n",
    "- cpichange: Imputaremos por iterative\n",
    "- grosstertiaryeducationenrollment: Imputaremos por knn\n",
    "- maternalmortalityratio: Imputaremos por knn"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "2:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "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>populationlaborforceparticipation</th>\n",
       "      <th>populationlaborforceparticipation_knn</th>\n",
       "      <th>populationlaborforceparticipation_iterative</th>\n",
       "      <th>totaltaxrate</th>\n",
       "      <th>totaltaxrate_knn</th>\n",
       "      <th>totaltaxrate_iterative</th>\n",
       "      <th>unemploymentrate</th>\n",
       "      <th>unemploymentrate_knn</th>\n",
       "      <th>unemploymentrate_iterative</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>176.000000</td>\n",
       "      <td>195.000000</td>\n",
       "      <td>195.000000</td>\n",
       "      <td>183.000000</td>\n",
       "      <td>195.000000</td>\n",
       "      <td>195.000000</td>\n",
       "      <td>176.000000</td>\n",
       "      <td>195.000000</td>\n",
       "      <td>195.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>62.738068</td>\n",
       "      <td>62.621562</td>\n",
       "      <td>62.675735</td>\n",
       "      <td>40.822404</td>\n",
       "      <td>40.791073</td>\n",
       "      <td>40.851151</td>\n",
       "      <td>6.886364</td>\n",
       "      <td>6.958282</td>\n",
       "      <td>6.883611</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>10.502907</td>\n",
       "      <td>10.023457</td>\n",
       "      <td>9.992136</td>\n",
       "      <td>20.557662</td>\n",
       "      <td>19.952092</td>\n",
       "      <td>19.980598</td>\n",
       "      <td>5.079216</td>\n",
       "      <td>4.869590</td>\n",
       "      <td>4.824268</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>38.000000</td>\n",
       "      <td>38.000000</td>\n",
       "      <td>38.000000</td>\n",
       "      <td>8.000000</td>\n",
       "      <td>8.000000</td>\n",
       "      <td>8.000000</td>\n",
       "      <td>0.090000</td>\n",
       "      <td>0.090000</td>\n",
       "      <td>0.090000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>56.150000</td>\n",
       "      <td>56.850000</td>\n",
       "      <td>57.250000</td>\n",
       "      <td>30.600000</td>\n",
       "      <td>30.750000</td>\n",
       "      <td>30.900000</td>\n",
       "      <td>3.395000</td>\n",
       "      <td>3.490000</td>\n",
       "      <td>3.590000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>62.450000</td>\n",
       "      <td>62.600000</td>\n",
       "      <td>62.675735</td>\n",
       "      <td>37.200000</td>\n",
       "      <td>37.700000</td>\n",
       "      <td>37.700000</td>\n",
       "      <td>5.360000</td>\n",
       "      <td>5.650000</td>\n",
       "      <td>5.920000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>69.500000</td>\n",
       "      <td>68.300000</td>\n",
       "      <td>68.300000</td>\n",
       "      <td>48.000000</td>\n",
       "      <td>47.600000</td>\n",
       "      <td>47.600000</td>\n",
       "      <td>9.490000</td>\n",
       "      <td>9.090000</td>\n",
       "      <td>8.840000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>86.800000</td>\n",
       "      <td>86.800000</td>\n",
       "      <td>86.800000</td>\n",
       "      <td>219.600000</td>\n",
       "      <td>219.600000</td>\n",
       "      <td>219.600000</td>\n",
       "      <td>28.180000</td>\n",
       "      <td>28.180000</td>\n",
       "      <td>28.180000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       populationlaborforceparticipation  \\\n",
       "count                         176.000000   \n",
       "mean                           62.738068   \n",
       "std                            10.502907   \n",
       "min                            38.000000   \n",
       "25%                            56.150000   \n",
       "50%                            62.450000   \n",
       "75%                            69.500000   \n",
       "max                            86.800000   \n",
       "\n",
       "       populationlaborforceparticipation_knn  \\\n",
       "count                             195.000000   \n",
       "mean                               62.621562   \n",
       "std                                10.023457   \n",
       "min                                38.000000   \n",
       "25%                                56.850000   \n",
       "50%                                62.600000   \n",
       "75%                                68.300000   \n",
       "max                                86.800000   \n",
       "\n",
       "       populationlaborforceparticipation_iterative  totaltaxrate  \\\n",
       "count                                   195.000000    183.000000   \n",
       "mean                                     62.675735     40.822404   \n",
       "std                                       9.992136     20.557662   \n",
       "min                                      38.000000      8.000000   \n",
       "25%                                      57.250000     30.600000   \n",
       "50%                                      62.675735     37.200000   \n",
       "75%                                      68.300000     48.000000   \n",
       "max                                      86.800000    219.600000   \n",
       "\n",
       "       totaltaxrate_knn  totaltaxrate_iterative  unemploymentrate  \\\n",
       "count        195.000000              195.000000        176.000000   \n",
       "mean          40.791073               40.851151          6.886364   \n",
       "std           19.952092               19.980598          5.079216   \n",
       "min            8.000000                8.000000          0.090000   \n",
       "25%           30.750000               30.900000          3.395000   \n",
       "50%           37.700000               37.700000          5.360000   \n",
       "75%           47.600000               47.600000          9.490000   \n",
       "max          219.600000              219.600000         28.180000   \n",
       "\n",
       "       unemploymentrate_knn  unemploymentrate_iterative  \n",
       "count            195.000000                  195.000000  \n",
       "mean               6.958282                    6.883611  \n",
       "std                4.869590                    4.824268  \n",
       "min                0.090000                    0.090000  \n",
       "25%                3.490000                    3.590000  \n",
       "50%                5.650000                    5.920000  \n",
       "75%                9.090000                    8.840000  \n",
       "max               28.180000                   28.180000  "
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_world_copia.describe()[['populationlaborforceparticipation','populationlaborforceparticipation_knn','populationlaborforceparticipation_iterative','totaltaxrate','totaltaxrate_knn','totaltaxrate_iterative', 'unemploymentrate', 'unemploymentrate_knn', 'unemploymentrate_iterative']]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "populationlaborforceparticipation: imputaremos por knn\n",
    "totaltaxrate: imputaremos por iterative\n",
    "unemploymentrate: imputaremos por  knn"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "#eliminamos las columnas que ya no nos interesan para guardar el DataFrame \n",
    "df_world_copia.drop(['cpi','cpi_knn', 'cpichange', 'cpichange_knn', 'grosstertiaryeducationenrollment','grosstertiaryeducationenrollment_iterative', 'maternalmortalityratio', 'maternalmortalityratio_iterative','populationlaborforceparticipation','populationlaborforceparticipation_iterative','totaltaxrate','totaltaxrate_knn', 'unemploymentrate', 'unemploymentrate_iterative' ], axis = 1, inplace = True)\n",
    "\n",
    "# ahora vamos a cambiar el nombre de las columnas que quedaron para que tengan el mismo nombre de origen\n",
    "nuevo_nombre = {'cpi_iterative': 'cpi',  'cpichange_iterative': 'cpichange', 'grosstertiaryeducationenrollment_knn':'grosstertiaryeducationenrollment', 'maternalmortalityratio_knn': 'maternalmortalityratio', 'populationlaborforceparticipation_knn': 'populationlaborforceparticipation', 'totaltaxrate_iterative':'totaltaxrate', 'unemploymentrate_knn': 'unemploymentrate'  }\n",
    "df_world_copia.rename(columns = nuevo_nombre, inplace = True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "________\n",
    "3 - Miramos los valores atípicos con el diagrama de boxplot:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1080x360 with 3 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "columnas_atipicas = ['armedforcessize', 'gasolineprice','taxrevenue' ]\n",
    "\n",
    "fig, axes = plt.subplots(nrows = 1, ncols = 4, figsize = (15,5)) \n",
    "\n",
    "axes = axes.flat\n",
    "\n",
    "for indice, col in enumerate(columnas_atipicas):\n",
    "    sns.boxplot(x = col, data = df_world_copia, ax = axes[indice])\n",
    "    \n",
    "plt.tight_layout()\n",
    "fig.delaxes(axes[-1]);"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Vemos que \"armed forced size\" es la columna que muestra más valores atípicos. \n",
    "\n",
    "1 - En éste caso, descartamos gestionar los nulos con la media, mediana o la moda. Lo pondremos como \"Unknown\"\n",
    "\n",
    "2 - Las columnas gasolineprice y taxrevenue las analizaremos pcon KNN Imputer y Iterative Imputer para valorar si se pueden imputar con estos métodos. \n",
    "\n",
    "\n",
    "_________\n",
    "\n",
    "1: ITERATIVE IMPUTER"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Después del reemplazo usando 'fillna' quedan los siguientes nulos\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "0"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 1: \n",
    "\n",
    "# reemplazamos los nulos por la categoría Unknown para cada una de las columnas de la lista\n",
    "df_world_copia['armedforcessize'] = df_world[columna].fillna(\"Unknown\")\n",
    "\n",
    "# comprobamos si quedan nulos en la columnas. \n",
    "print(\"Después del reemplazo usando 'fillna' quedan los siguientes nulos\")\n",
    "df_world_copia['armedforcessize'].isnull().sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "2: KNN IMPUTER:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/lluerna/opt/anaconda3/lib/python3.9/site-packages/sklearn/utils/validation.py:571: FutureWarning: is_sparse is deprecated and will be removed in a future version. Check `isinstance(dtype, pd.SparseDtype)` instead.\n",
      "  array.dtypes.apply(is_sparse).any()):\n",
      "/Users/lluerna/opt/anaconda3/lib/python3.9/site-packages/sklearn/utils/validation.py:571: FutureWarning: is_sparse is deprecated and will be removed in a future version. Check `isinstance(dtype, pd.SparseDtype)` instead.\n",
      "  array.dtypes.apply(is_sparse).any()):\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "array([[ 0.7       ,  9.3       ],\n",
       "       [ 1.36      , 18.6       ],\n",
       "       [ 0.28      , 37.2       ],\n",
       "       [ 1.51      , 22.44      ],\n",
       "       [ 0.97      ,  9.2       ],\n",
       "       [ 0.99      , 16.5       ],\n",
       "       [ 1.1       , 10.1       ],\n",
       "       [ 0.77      , 20.9       ],\n",
       "       [ 0.93      , 23.        ],\n",
       "       [ 1.2       , 25.4       ],\n",
       "       [ 0.56      , 13.        ],\n",
       "       [ 0.92      , 14.8       ],\n",
       "       [ 0.43      ,  4.2       ],\n",
       "       [ 1.12      ,  8.8       ],\n",
       "       [ 1.81      , 27.5       ],\n",
       "       [ 0.6       , 14.7       ],\n",
       "       [ 1.43      , 24.        ],\n",
       "       [ 1.13      , 26.3       ],\n",
       "       [ 0.72      , 10.8       ],\n",
       "       [ 0.98      , 16.        ],\n",
       "       [ 0.71      , 17.        ],\n",
       "       [ 1.05      , 20.4       ],\n",
       "       [ 0.71      , 19.5       ],\n",
       "       [ 1.02      , 14.2       ],\n",
       "       [ 0.37      ,  9.54      ],\n",
       "       [ 1.11      , 20.2       ],\n",
       "       [ 0.98      , 15.        ],\n",
       "       [ 1.21      , 13.6       ],\n",
       "       [ 0.93      , 11.8       ],\n",
       "       [ 1.02      , 20.1       ],\n",
       "       [ 0.9       , 17.1       ],\n",
       "       [ 1.03      , 12.8       ],\n",
       "       [ 0.81      , 12.8       ],\n",
       "       [ 1.41      ,  8.6       ],\n",
       "       [ 0.78      , 15.06      ],\n",
       "       [ 1.03      , 18.2       ],\n",
       "       [ 0.96      ,  9.4       ],\n",
       "       [ 0.68      , 14.4       ],\n",
       "       [ 1.00245714, 16.5739645 ],\n",
       "       [ 0.97      ,  9.        ],\n",
       "       [ 0.98      , 13.6       ],\n",
       "       [ 1.26      , 22.        ],\n",
       "       [ 1.4       , 16.18      ],\n",
       "       [ 1.23      , 24.5       ],\n",
       "       [ 1.17      , 14.9       ],\n",
       "       [ 1.49      , 10.7       ],\n",
       "       [ 1.55      , 32.4       ],\n",
       "       [ 1.32      , 18.8       ],\n",
       "       [ 1.142     , 22.1       ],\n",
       "       [ 1.07      , 13.        ],\n",
       "       [ 0.61      ,  7.66      ],\n",
       "       [ 0.4       , 12.5       ],\n",
       "       [ 0.83      , 18.1       ],\n",
       "       [ 0.614     ,  6.1       ],\n",
       "       [ 2.        , 23.44      ],\n",
       "       [ 1.14      , 20.9       ],\n",
       "       [ 1.262     , 28.6       ],\n",
       "       [ 0.75      ,  7.5       ],\n",
       "       [ 0.82      , 24.2       ],\n",
       "       [ 1.45      , 20.8       ],\n",
       "       [ 1.39      , 24.2       ],\n",
       "       [ 0.92      , 10.2       ],\n",
       "       [ 1.18      ,  9.4       ],\n",
       "       [ 0.76      , 21.7       ],\n",
       "       [ 1.39      , 11.5       ],\n",
       "       [ 0.92      , 12.6       ],\n",
       "       [ 1.54      , 26.2       ],\n",
       "       [ 1.12      , 19.4       ],\n",
       "       [ 0.79      , 10.6       ],\n",
       "       [ 0.9       , 10.8       ],\n",
       "       [ 1.042     , 10.3       ],\n",
       "       [ 0.9       , 17.94      ],\n",
       "       [ 0.81      , 17.34      ],\n",
       "       [ 1.00245714, 16.5739645 ],\n",
       "       [ 0.98      , 17.3       ],\n",
       "       [ 1.18      , 23.        ],\n",
       "       [ 1.69      , 23.3       ],\n",
       "       [ 0.97      , 11.2       ],\n",
       "       [ 0.63      , 10.2       ],\n",
       "       [ 0.4       ,  7.4       ],\n",
       "       [ 0.61      ,  2.        ],\n",
       "       [ 1.37      , 18.3       ],\n",
       "       [ 1.57      , 23.1       ],\n",
       "       [ 1.61      , 24.3       ],\n",
       "       [ 1.11      , 26.8       ],\n",
       "       [ 1.06      , 11.9       ],\n",
       "       [ 1.1       , 15.1       ],\n",
       "       [ 0.42      , 11.7       ],\n",
       "       [ 0.95      , 15.1       ],\n",
       "       [ 1.056     , 22.        ],\n",
       "       [ 0.35      ,  1.4       ],\n",
       "       [ 0.56      , 18.        ],\n",
       "       [ 0.93      , 12.9       ],\n",
       "       [ 1.16      , 22.9       ],\n",
       "       [ 0.74      , 15.3       ],\n",
       "       [ 0.7       , 31.6       ],\n",
       "       [ 0.8       , 12.9       ],\n",
       "       [ 0.11      , 13.48      ],\n",
       "       [ 1.74      , 23.44      ],\n",
       "       [ 1.16      , 16.9       ],\n",
       "       [ 1.19      , 26.5       ],\n",
       "       [ 1.11      , 10.2       ],\n",
       "       [ 1.15      , 17.3       ],\n",
       "       [ 0.45      , 12.        ],\n",
       "       [ 1.63      , 19.5       ],\n",
       "       [ 1.12      , 11.6       ],\n",
       "       [ 1.36      , 26.2       ],\n",
       "       [ 1.44      , 17.8       ],\n",
       "       [ 1.13      , 20.4       ],\n",
       "       [ 1.12      , 19.1       ],\n",
       "       [ 0.73      , 13.1       ],\n",
       "       [ 1.18      , 25.2       ],\n",
       "       [ 0.8       , 17.7       ],\n",
       "       [ 2.        , 23.44      ],\n",
       "       [ 0.72      , 16.8       ],\n",
       "       [ 1.16      , 18.06      ],\n",
       "       [ 0.99      , 21.9       ],\n",
       "       [ 0.65      ,  0.        ],\n",
       "       [ 0.54      ,  5.4       ],\n",
       "       [ 0.76      , 27.1       ],\n",
       "       [ 1.00245714, 16.5739645 ],\n",
       "       [ 0.91      , 20.7       ],\n",
       "       [ 1.68      , 23.        ],\n",
       "       [ 1.4       , 29.        ],\n",
       "       [ 0.91      , 15.6       ],\n",
       "       [ 0.88      , 11.8       ],\n",
       "       [ 0.46      ,  1.5       ],\n",
       "       [ 0.58      , 11.82      ],\n",
       "       [ 1.00245714, 16.5739645 ],\n",
       "       [ 1.78      , 23.9       ],\n",
       "       [ 0.45      ,  2.5       ],\n",
       "       [ 0.79      ,  9.2       ],\n",
       "       [ 0.97      , 21.3       ],\n",
       "       [ 1.00245714, 16.5739645 ],\n",
       "       [ 0.74      , 13.56      ],\n",
       "       [ 1.36      , 13.6       ],\n",
       "       [ 1.04      , 10.        ],\n",
       "       [ 0.99      , 14.3       ],\n",
       "       [ 0.86      , 14.        ],\n",
       "       [ 1.07      , 17.4       ],\n",
       "       [ 1.54      , 22.8       ],\n",
       "       [ 0.4       , 14.7       ],\n",
       "       [ 1.16      , 14.6       ],\n",
       "       [ 0.59      , 11.4       ],\n",
       "       [ 1.17      , 14.3       ],\n",
       "       [ 1.306     , 18.5       ],\n",
       "       [ 1.3       , 18.2       ],\n",
       "       [ 1.242     , 25.4       ],\n",
       "       [ 0.91      , 25.5       ],\n",
       "       [ 1.028     , 18.1       ],\n",
       "       [ 0.752     , 14.6       ],\n",
       "       [ 0.24      ,  8.9       ],\n",
       "       [ 1.14      , 16.3       ],\n",
       "       [ 1.16      , 18.6       ],\n",
       "       [ 1.07      , 34.1       ],\n",
       "       [ 1.08      ,  8.6       ],\n",
       "       [ 1.25      , 13.1       ],\n",
       "       [ 1.32      , 18.7       ],\n",
       "       [ 1.32      , 18.6       ],\n",
       "       [ 1.25      , 29.5       ],\n",
       "       [ 1.41      ,  0.        ],\n",
       "       [ 0.92      , 27.5       ],\n",
       "       [ 1.22      , 15.6       ],\n",
       "       [ 0.28      , 14.94      ],\n",
       "       [ 1.26      , 14.2       ],\n",
       "       [ 0.88      , 11.9       ],\n",
       "       [ 0.95      ,  8.        ],\n",
       "       [ 1.29      , 19.5       ],\n",
       "       [ 1.42      , 27.9       ],\n",
       "       [ 1.45      , 10.1       ],\n",
       "       [ 0.83      , 14.2       ],\n",
       "       [ 0.71      ,  9.8       ],\n",
       "       [ 0.87      , 11.5       ],\n",
       "       [ 0.71      , 14.9       ],\n",
       "       [ 1.1       , 25.        ],\n",
       "       [ 0.71      , 16.9       ],\n",
       "       [ 1.142     , 22.3       ],\n",
       "       [ 0.54      , 19.5       ],\n",
       "       [ 0.73      , 21.1       ],\n",
       "       [ 1.42      , 17.9       ],\n",
       "       [ 0.29      , 13.48      ],\n",
       "       [ 1.00245714, 16.5739645 ],\n",
       "       [ 0.94      , 11.7       ],\n",
       "       [ 0.83      , 20.1       ],\n",
       "       [ 0.49      ,  0.1       ],\n",
       "       [ 1.46      , 25.5       ],\n",
       "       [ 0.71      ,  9.6       ],\n",
       "       [ 1.5       , 20.1       ],\n",
       "       [ 1.03      , 14.8       ],\n",
       "       [ 1.31      , 17.8       ],\n",
       "       [ 0.        , 14.94      ],\n",
       "       [ 0.8       , 19.1       ],\n",
       "       [ 0.92      , 18.12      ],\n",
       "       [ 1.4       , 16.2       ],\n",
       "       [ 1.34      , 20.7       ]])"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# instanciamos la clase del KNNImputer\n",
    "imputer_knn_2 = KNNImputer(n_neighbors = 5)\n",
    "\n",
    "# ajustamos y transformamos los datos\n",
    "imputer_knn_imputado_2 = imputer_knn.fit_transform(df_world_copia[['gasolineprice','taxrevenue']])\n",
    "\n",
    "# comprobamos que es lo que nos devuelve, que sigue siendo un array\n",
    "imputer_knn_imputado_2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Después del 'KNN' tenemos: \n",
      "gasolineprice_knn    0\n",
      "taxrevenue_knn       0\n",
      "dtype: int64 nulos\n"
     ]
    }
   ],
   "source": [
    "# por último nos queda añadir ese array al DataFrame como hemos hecho hasta ahora\n",
    "df_world_copia[['gasolineprice_knn','taxrevenue_knn']] = imputer_knn_imputado_2\n",
    "\n",
    "\n",
    "# comprobamos los nulos\n",
    "print(f\"Después del 'KNN' tenemos: \\n{df_world_copia[['gasolineprice_knn','taxrevenue_knn']].isnull().sum()} nulos\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "IterativeImputer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/lluerna/opt/anaconda3/lib/python3.9/site-packages/sklearn/utils/validation.py:571: FutureWarning: is_sparse is deprecated and will be removed in a future version. Check `isinstance(dtype, pd.SparseDtype)` instead.\n",
      "  array.dtypes.apply(is_sparse).any()):\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "array([[ 0.7       ,  9.3       ],\n",
       "       [ 1.36      , 18.6       ],\n",
       "       [ 0.28      , 37.2       ],\n",
       "       [ 1.51      , 21.32146501],\n",
       "       [ 0.97      ,  9.2       ],\n",
       "       [ 0.99      , 16.5       ],\n",
       "       [ 1.1       , 10.1       ],\n",
       "       [ 0.77      , 20.9       ],\n",
       "       [ 0.93      , 23.        ],\n",
       "       [ 1.2       , 25.4       ],\n",
       "       [ 0.56      , 13.        ],\n",
       "       [ 0.92      , 14.8       ],\n",
       "       [ 0.43      ,  4.2       ],\n",
       "       [ 1.12      ,  8.8       ],\n",
       "       [ 1.81      , 27.5       ],\n",
       "       [ 0.6       , 14.7       ],\n",
       "       [ 1.43      , 24.        ],\n",
       "       [ 1.13      , 26.3       ],\n",
       "       [ 0.72      , 10.8       ],\n",
       "       [ 0.98      , 16.        ],\n",
       "       [ 0.71      , 17.        ],\n",
       "       [ 1.05      , 20.4       ],\n",
       "       [ 0.71      , 19.5       ],\n",
       "       [ 1.02      , 14.2       ],\n",
       "       [ 0.37      , 10.22179113],\n",
       "       [ 1.11      , 20.2       ],\n",
       "       [ 0.98      , 15.        ],\n",
       "       [ 1.21      , 13.6       ],\n",
       "       [ 0.93      , 11.8       ],\n",
       "       [ 1.02      , 20.1       ],\n",
       "       [ 0.9       , 17.1       ],\n",
       "       [ 1.03      , 12.8       ],\n",
       "       [ 0.81      , 12.8       ],\n",
       "       [ 1.41      ,  8.6       ],\n",
       "       [ 0.78      , 14.2137791 ],\n",
       "       [ 1.03      , 18.2       ],\n",
       "       [ 0.96      ,  9.4       ],\n",
       "       [ 0.68      , 14.4       ],\n",
       "       [ 1.01289083, 16.48133375],\n",
       "       [ 0.97      ,  9.        ],\n",
       "       [ 0.98      , 13.6       ],\n",
       "       [ 1.26      , 22.        ],\n",
       "       [ 1.4       , 20.25044385],\n",
       "       [ 1.23      , 24.5       ],\n",
       "       [ 1.17      , 14.9       ],\n",
       "       [ 1.49      , 10.7       ],\n",
       "       [ 1.55      , 32.4       ],\n",
       "       [ 1.32      , 19.47151936],\n",
       "       [ 1.16519085, 22.1       ],\n",
       "       [ 1.07      , 13.        ],\n",
       "       [ 0.61      , 12.55856458],\n",
       "       [ 0.4       , 12.5       ],\n",
       "       [ 0.83      , 18.1       ],\n",
       "       [ 0.73148535,  6.1       ],\n",
       "       [ 2.        , 26.09237747],\n",
       "       [ 1.14      , 20.9       ],\n",
       "       [ 1.34138371, 28.6       ],\n",
       "       [ 0.75      ,  7.5       ],\n",
       "       [ 0.82      , 24.2       ],\n",
       "       [ 1.45      , 20.8       ],\n",
       "       [ 1.39      , 24.2       ],\n",
       "       [ 0.92      , 10.2       ],\n",
       "       [ 1.18      ,  9.4       ],\n",
       "       [ 0.76      , 21.7       ],\n",
       "       [ 1.39      , 11.5       ],\n",
       "       [ 0.92      , 12.6       ],\n",
       "       [ 1.54      , 26.2       ],\n",
       "       [ 1.12      , 19.4       ],\n",
       "       [ 0.79      , 10.6       ],\n",
       "       [ 0.9       , 10.8       ],\n",
       "       [ 0.84533304, 10.3       ],\n",
       "       [ 0.9       , 15.38216583],\n",
       "       [ 0.81      , 14.50587579],\n",
       "       [ 1.01289083, 16.48133375],\n",
       "       [ 0.98      , 17.3       ],\n",
       "       [ 1.18      , 23.        ],\n",
       "       [ 1.69      , 23.3       ],\n",
       "       [ 0.97      , 11.2       ],\n",
       "       [ 0.63      , 10.2       ],\n",
       "       [ 0.4       ,  7.4       ],\n",
       "       [ 0.61      ,  2.        ],\n",
       "       [ 1.37      , 18.3       ],\n",
       "       [ 1.57      , 23.1       ],\n",
       "       [ 1.61      , 24.3       ],\n",
       "       [ 1.11      , 26.8       ],\n",
       "       [ 1.06      , 11.9       ],\n",
       "       [ 1.1       , 15.1       ],\n",
       "       [ 0.42      , 11.7       ],\n",
       "       [ 0.95      , 15.1       ],\n",
       "       [ 1.16248019, 22.        ],\n",
       "       [ 0.35      ,  1.4       ],\n",
       "       [ 0.56      , 18.        ],\n",
       "       [ 0.93      , 12.9       ],\n",
       "       [ 1.16      , 22.9       ],\n",
       "       [ 0.74      , 15.3       ],\n",
       "       [ 0.7       , 31.6       ],\n",
       "       [ 0.8       , 12.9       ],\n",
       "       [ 0.11      ,  7.69028656],\n",
       "       [ 1.74      , 23.5608729 ],\n",
       "       [ 1.16      , 16.9       ],\n",
       "       [ 1.19      , 26.5       ],\n",
       "       [ 1.11      , 10.2       ],\n",
       "       [ 1.15      , 17.3       ],\n",
       "       [ 0.45      , 12.        ],\n",
       "       [ 1.63      , 19.5       ],\n",
       "       [ 1.12      , 11.6       ],\n",
       "       [ 1.36      , 26.2       ],\n",
       "       [ 1.44      , 17.8       ],\n",
       "       [ 1.13      , 17.62157372],\n",
       "       [ 1.12      , 19.1       ],\n",
       "       [ 0.73      , 13.1       ],\n",
       "       [ 1.24922129, 25.2       ],\n",
       "       [ 0.8       , 17.7       ],\n",
       "       [ 2.        , 26.09237747],\n",
       "       [ 0.72      , 16.8       ],\n",
       "       [ 1.16      , 17.9136704 ],\n",
       "       [ 0.99      , 21.9       ],\n",
       "       [ 0.65      ,  0.        ],\n",
       "       [ 0.54      ,  5.4       ],\n",
       "       [ 0.76      , 27.1       ],\n",
       "       [ 1.01289083, 16.48133375],\n",
       "       [ 0.91      , 20.7       ],\n",
       "       [ 1.68      , 23.        ],\n",
       "       [ 1.4       , 29.        ],\n",
       "       [ 0.91      , 15.6       ],\n",
       "       [ 0.88      , 11.8       ],\n",
       "       [ 0.46      ,  1.5       ],\n",
       "       [ 0.58      , 12.2664679 ],\n",
       "       [ 1.01289083, 16.48133375],\n",
       "       [ 1.78      , 23.9       ],\n",
       "       [ 0.45      ,  2.5       ],\n",
       "       [ 0.79      ,  9.2       ],\n",
       "       [ 1.14350557, 21.3       ],\n",
       "       [ 1.01289083, 16.48133375],\n",
       "       [ 0.74      , 13.82431686],\n",
       "       [ 1.36      , 13.6       ],\n",
       "       [ 1.04      , 10.        ],\n",
       "       [ 0.99      , 14.3       ],\n",
       "       [ 0.86      , 14.        ],\n",
       "       [ 1.07      , 17.4       ],\n",
       "       [ 1.54      , 22.8       ],\n",
       "       [ 0.4       , 14.7       ],\n",
       "       [ 1.16      , 14.6       ],\n",
       "       [ 0.59      , 11.4       ],\n",
       "       [ 1.17      , 14.3       ],\n",
       "       [ 1.06760711, 18.5       ],\n",
       "       [ 1.3       , 18.2       ],\n",
       "       [ 1.25464261, 25.4       ],\n",
       "       [ 0.91      , 25.5       ],\n",
       "       [ 1.05676447, 18.1       ],\n",
       "       [ 0.96189139, 14.6       ],\n",
       "       [ 0.24      ,  8.9       ],\n",
       "       [ 1.14      , 16.3       ],\n",
       "       [ 1.16      , 18.6       ],\n",
       "       [ 1.49046997, 34.1       ],\n",
       "       [ 1.08      ,  8.6       ],\n",
       "       [ 1.25      , 13.1       ],\n",
       "       [ 1.32      , 18.7       ],\n",
       "       [ 1.32      , 18.6       ],\n",
       "       [ 1.36577964, 29.5       ],\n",
       "       [ 1.41      ,  0.        ],\n",
       "       [ 0.92      , 27.5       ],\n",
       "       [ 1.22      , 15.6       ],\n",
       "       [ 0.28      ,  9.34550108],\n",
       "       [ 1.26      , 14.2       ],\n",
       "       [ 0.88      , 11.9       ],\n",
       "       [ 0.95      ,  8.        ],\n",
       "       [ 1.29      , 19.5       ],\n",
       "       [ 1.42      , 27.9       ],\n",
       "       [ 1.45      , 10.1       ],\n",
       "       [ 0.83      , 14.2       ],\n",
       "       [ 0.71      ,  9.8       ],\n",
       "       [ 0.87      , 11.5       ],\n",
       "       [ 0.71      , 14.9       ],\n",
       "       [ 1.1       , 25.        ],\n",
       "       [ 0.71      , 16.9       ],\n",
       "       [ 1.17061217, 22.3       ],\n",
       "       [ 0.54      , 19.5       ],\n",
       "       [ 0.73      , 21.1       ],\n",
       "       [ 1.42      , 17.9       ],\n",
       "       [ 0.29      ,  9.44286665],\n",
       "       [ 1.01289083, 16.48133375],\n",
       "       [ 0.94      , 11.7       ],\n",
       "       [ 0.83      , 20.1       ],\n",
       "       [ 0.49      ,  0.1       ],\n",
       "       [ 1.46      , 25.5       ],\n",
       "       [ 0.71      ,  9.6       ],\n",
       "       [ 1.5       , 20.1       ],\n",
       "       [ 1.03      , 14.8       ],\n",
       "       [ 1.31      , 17.8       ],\n",
       "       [ 0.        ,  6.61926539],\n",
       "       [ 0.8       , 19.1       ],\n",
       "       [ 0.92      , 15.57689695],\n",
       "       [ 1.4       , 16.2       ],\n",
       "       [ 1.34      , 20.7       ]])"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# instanciamos las clases\n",
    "imputer_iterative_2 = IterativeImputer(max_iter = 20, random_state = 42)\n",
    "\n",
    "# ajustamos y tranformamos los datos\n",
    "imputer_iterative_imputado_2 = imputer_iterative_2.fit_transform(df_world_copia[['gasolineprice','taxrevenue']])\n",
    "\n",
    "# comprobamos que es lo que nos devuelve, que en este caso es un array también\n",
    "imputer_iterative_imputado_2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Después del 'Iterative' tenemos: \n",
      "gasolineprice_iterative    0\n",
      "taxrevenue_iterative       0\n",
      "dtype: int64 nulos\n"
     ]
    }
   ],
   "source": [
    "# en este caso como vamos a aplicar los dos métodos sobre estas columnas vamos a crear dos columnas nuevas\n",
    "df_world_copia[['gasolineprice_iterative','taxrevenue_iterative']] = imputer_iterative_imputado_2\n",
    "\n",
    "# comprobamos los nulos\n",
    "print(f\"Después del 'Iterative' tenemos: \\n{df_world_copia[['gasolineprice_iterative','taxrevenue_iterative']].isnull().sum()} nulos\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Comparamos los datos"
   ]
  },
  {
   "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></th>\n",
       "      <th>gasolineprice</th>\n",
       "      <th>gasolineprice_knn</th>\n",
       "      <th>gasolineprice_iterative</th>\n",
       "      <th>taxrevenue</th>\n",
       "      <th>taxrevenue_knn</th>\n",
       "      <th>taxrevenue_iterative</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>175.000000</td>\n",
       "      <td>195.000000</td>\n",
       "      <td>195.000000</td>\n",
       "      <td>169.000000</td>\n",
       "      <td>195.000000</td>\n",
       "      <td>195.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>1.002457</td>\n",
       "      <td>1.007696</td>\n",
       "      <td>1.012891</td>\n",
       "      <td>16.573964</td>\n",
       "      <td>16.587301</td>\n",
       "      <td>16.481334</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>0.368858</td>\n",
       "      <td>0.353491</td>\n",
       "      <td>0.355154</td>\n",
       "      <td>6.961632</td>\n",
       "      <td>6.631697</td>\n",
       "      <td>6.732639</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>0.755000</td>\n",
       "      <td>0.775000</td>\n",
       "      <td>0.785000</td>\n",
       "      <td>11.700000</td>\n",
       "      <td>11.900000</td>\n",
       "      <td>11.800000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>0.980000</td>\n",
       "      <td>1.002457</td>\n",
       "      <td>1.012891</td>\n",
       "      <td>16.300000</td>\n",
       "      <td>16.573964</td>\n",
       "      <td>16.481334</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>1.240000</td>\n",
       "      <td>1.225000</td>\n",
       "      <td>1.239611</td>\n",
       "      <td>21.100000</td>\n",
       "      <td>20.850000</td>\n",
       "      <td>20.850000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>2.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>37.200000</td>\n",
       "      <td>37.200000</td>\n",
       "      <td>37.200000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       gasolineprice  gasolineprice_knn  gasolineprice_iterative  taxrevenue  \\\n",
       "count     175.000000         195.000000               195.000000  169.000000   \n",
       "mean        1.002457           1.007696                 1.012891   16.573964   \n",
       "std         0.368858           0.353491                 0.355154    6.961632   \n",
       "min         0.000000           0.000000                 0.000000    0.000000   \n",
       "25%         0.755000           0.775000                 0.785000   11.700000   \n",
       "50%         0.980000           1.002457                 1.012891   16.300000   \n",
       "75%         1.240000           1.225000                 1.239611   21.100000   \n",
       "max         2.000000           2.000000                 2.000000   37.200000   \n",
       "\n",
       "       taxrevenue_knn  taxrevenue_iterative  \n",
       "count      195.000000            195.000000  \n",
       "mean        16.587301             16.481334  \n",
       "std          6.631697              6.732639  \n",
       "min          0.000000              0.000000  \n",
       "25%         11.900000             11.800000  \n",
       "50%         16.573964             16.481334  \n",
       "75%         20.850000             20.850000  \n",
       "max         37.200000             37.200000  "
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_world_copia.describe()[['gasolineprice','gasolineprice_knn','gasolineprice_iterative','taxrevenue', 'taxrevenue_knn', 'taxrevenue_iterative']]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "gasolineprice = imputamos por knn\n",
    "\n",
    "taxrevenue = imputamos por knn\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [],
   "source": [
    "#eliminamos las columnas que ya no nos interesan para guardar el DataFrame \n",
    "df_world_copia.drop(['gasolineprice','gasolineprice_iterative','taxrevenue', 'taxrevenue_iterative' ], axis = 1, inplace = True)\n",
    "\n",
    "# ahora vamos a cambiar el nombre de las columnas que quedaron para que tengan el mismo nombre de origen\n",
    "nuevo_nombre = {'gasolineprice_knn': 'gasolineprice',  'taxrevenue_knn': 'taxrevenue' }\n",
    "df_world_copia.rename(columns = nuevo_nombre, inplace = True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "country                              0\n",
       "density                              0\n",
       "abbreviation                         0\n",
       "agriculturalland                     0\n",
       "landarea                             0\n",
       "armedforcessize                      0\n",
       "birthrate                            0\n",
       "callingcode                          0\n",
       "capital/majorcity                    0\n",
       "co2-emissions                        0\n",
       "currency-code                        0\n",
       "fertilityrate                        0\n",
       "forestedarea                         0\n",
       "gdp                                  0\n",
       "grossprimaryeducationenrollment      0\n",
       "infantmortality                      0\n",
       "largestcity                          0\n",
       "lifeexpectancy                       0\n",
       "minimumwage                          0\n",
       "officiallanguage                     0\n",
       "outofpockethealthexpenditure         0\n",
       "physiciansperthousand                0\n",
       "population                           0\n",
       "urban_population                     0\n",
       "longitud                             0\n",
       "latitud                              0\n",
       "continent                            0\n",
       "cpi                                  0\n",
       "cpichange                            0\n",
       "totaltaxrate                         0\n",
       "grosstertiaryeducationenrollment     0\n",
       "maternalmortalityratio               0\n",
       "populationlaborforceparticipation    0\n",
       "unemploymentrate                     0\n",
       "gasolineprice                        0\n",
       "taxrevenue                           0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#comprobamos que no tenemos nulos:\n",
    "\n",
    "df_world_copia.isnull().sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Por último, pasamos el Dataframe a .csv:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_world_copia.to_csv(\"merge_world_nulos_gestionados.csv\",index = False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "_______________\n",
    "\n",
    "# Resumen de cómo se han gestionado los nulos:\n",
    "\n",
    "1 - Hemos mirado cuántas columnas tenían nulos: un total de 34 columnas, casi todas, ya que en total hay 36.\n",
    "\n",
    "2 - En un primer análisis de los datos vemos:\n",
    "- La columna \"minimum wage\" con un 23% de datos. Decidimos mantener la columna. Pasamos los nulos a la categoría \"Unknow\", ya que es un porcentaje elevado y podría dar datos alejados de la realidad.\n",
    "\n",
    "Seguimos analizando, por un lado los valores categóricos y por otro los numéricos.\n",
    "\n",
    "**Valores categóricos:**\n",
    "\n",
    "Las columnas categóricas que tienen nulos son : \n",
    "'abbreviation', 'capital/majorcity', 'currency-code', 'largestcity' 'officiallanguage', 'continent'\n",
    "\n",
    "- No tiene sentido substituir estos valores por la media o por la moda, por lo que creamos la categotía \"Unknow\" para clasificar todos estos valores nulos\n",
    "\n",
    "**Valores numéricos**\n",
    "\n",
    "Las columnas numéricas que tienen nulos son : \n",
    "'agriculturalland', 'landarea', 'armedforcessize', 'birthrate',\n",
    "       'callingcode', 'co2-emissions', 'cpi', 'cpichange', 'fertilityrate',\n",
    "       'forestedarea', 'gasolineprice', 'gdp',\n",
    "       'grossprimaryeducationenrollment', 'grosstertiaryeducationenrollment',\n",
    "       'infantmortality', 'lifeexpectancy', 'maternalmortalityratio',\n",
    "       'outofpockethealthexpenditure', 'physiciansperthousand', 'population',\n",
    "       'populationlaborforceparticipation', 'taxrevenue', 'totaltaxrate',\n",
    "       'unemploymentrate', 'urban_population', 'longitud', 'latitud'\n",
    "\n",
    "**Criterio para gestionar los nulos de las variables numéricas:**\n",
    "\n",
    "1 - Las columnas con porcentajes inferiores o igual al 5% se imputan por la media.\n",
    "\n",
    "2- Aquellas columnas que tienen valores entre el 5-10% podemos optar por reemplazar por la media o mediana. Haremos una comparativa para valorar si se imputa con KNN Imputer o co Iterative Imputer. Se escoge el método que más se asemeja al dato original. \n",
    "\n",
    "3- Hay tres columnas ('armedforcessize', 'gasolineprice','taxrevenue' ) con valores superiores o igual al 10%: \n",
    "- armedforcessize: vemos que hay muchos valores atípicos, por lo que los nulos pasan a la categoría Unknown\n",
    "- gasolineprice y taxrevenue: comparamos los resultados de imputar knn y iterative, y los imputamos en knn. Aunque hay muy poca variación entre un método y otro."
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "base",
   "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": "3.9.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}