{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Teorema Central do Limite"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "por Danilo J. S. Bellini"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "O objetivo dos códigos neste notebook é possibilitar a ilustração/aplicação do conteúdo matemático/estatístico, de forma que o leitor não precisa se prender aos mesmos para a compreensão do conteúdo exposto."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "O Teorema do Limite Central é um resultado fundamental em teoria de probabilidades e estatística. Resumidamente, o teorema diz que a média e a soma de variáveis aleatórias independentes identicamente distribuídas (*i.i.d.*) convergem em distribuição para a distribuição normal, e que isso (até certo ponto) independe da distribuição original."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "from scipy import stats\n",
    "from sympy import Symbol, I, exp\n",
    "from sympy.stats import Normal, E\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "np.random.seed(42)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "t = Symbol(\"t\", positive=True)\n",
    "mu = Symbol(\"mu\", real=True)\n",
    "sigma = Symbol(\"sigma\", positive=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Introdução: RPG - Role Playing Game com dados!"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "![](my_die.jpg)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Estamos jogando RPG! Os resultados das ações do nosso personagem dependem do resultado de lançamentos de dados como $d4$, $d10$ e $d20$, em que o número representa a quantidade de lados do dado. A notação $KdN$ representa a soma de $K$ lançamentos do dado $dN$.\n",
    "\n",
    "Durante uma batalha, você deve lançar um dado de $20$ lados ($1d20$). Nas condições dessa batalha, se seu resultado for maior ou igual a $16$, você pode vencer, do contrário, seu personagem poderá será derrotado.\n",
    "\n",
    "Diante dessa situação, o mestre do RPG sugere uma alternativa: lançar $2$ vezes o $d10$ e utilizar a soma como se esta fosse um lançamento do $d20$. Ou mesmo lançar $5$ vezes o $d4$. Admitindo que os dados não sejam viciados, qual abordagem maximiza a probabilidade de termos um resultado maior ou igual a $16$, com $1d20$, $2d10$ ou $5d4$?"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Para um dado convencional de $10$ lados não-viciado, temos a seguinte distribuição de probabilidades:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAX0AAAEKCAYAAAD+XoUoAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f50244ac208>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sides = range(1, 11)\n",
    "plt.bar(sides, .1)\n",
    "plt.title(\"Função massa de probabilidades do $d10$\");"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Em um experimento, o histograma dificilmente será idêntico a essa distribuição. Por exemplo, simulando $2000$ lançamentos desse dado, temos:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXoAAAEKCAYAAAAcgp5RAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nuP8Dc+mJFCGstQEAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f4ff4ea7a20>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "events10 = np.random.randint(1, 11, size=2000)\n",
    "plt.bar(sides, np.bincount(events10)[1:])\n",
    "plt.title(\"Histograma de $2000$ lançamentos do $d10$\");"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Se a ideia é utilizar $2d10$ ao invés de $1d20$, podemos pegar pares adjacentes desses lançamentos e utilizar as somas como nossos *compounds*. O resultado é:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXoAAAEKCAYAAAAcgp5RAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f4ff4e95c18>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.bar(range(2, 21), np.bincount(events10.reshape(-1, 2).sum(axis=1))[2:])\n",
    "plt.title(\"Histograma de $1000$ eventos $2d10$\");"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Embora o espaço amostral da composição $2d10$ seja menor que o espaço amostral do evento $1d20$ devido à impossibilidade de se obter o valor $1$ com o resultado da soma de $2$ lançamentos, a distribuição resultante dos $2$ lançamentos é triangular, o que torna o eventos como o $20$ improvável."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Com $5$ lançamentos de dados de $4$ lados, obteríamos:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXoAAAEKCAYAAAAcgp5RAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nzkFvZpY5B72ZWeb+PwGfob6V2BdEAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f4ff4dcc748>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "events4 = np.random.randint(1, 5, size=5000)\n",
    "plt.bar(range(5, 21), np.bincount(events4.reshape(-1, 5).sum(axis=1))[5:])\n",
    "plt.title(\"Histograma de $1000$ eventos $5d4$\");"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Esse novo histograma já não é mais um triângulo."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Utilizando as frequências relativas empiricamente coletadas como estimativa da probabilidade de obtermos um número maior ou igual a 16, teríamos:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "events20 = np.random.randint(1, 21, size=1000)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "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>$1d20$</th>\n",
       "      <th>$2d10$</th>\n",
       "      <th>$5d4$</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Frequências relativas em que $X \\ge 16$</th>\n",
       "      <td>0.263</td>\n",
       "      <td>0.147</td>\n",
       "      <td>0.122</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                         $1d20$  $2d10$  $5d4$\n",
       "Frequências relativas em que $X \\ge 16$   0.263   0.147  0.122"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "freqs_dframe = pd.DataFrame({\n",
    "    \"$1d20$\": [(events20 >= 16).mean()],\n",
    "    \"$2d10$\": [(events10.reshape(-1, 2).sum(axis=1) >= 16).mean()],\n",
    "    \"$5d4$\": [(events4.reshape(-1, 5).sum(axis=1) >= 16).mean()],\n",
    "}, index=[\"Frequências relativas em que $X \\ge 16$\"])\n",
    "freqs_dframe"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Aparentemente não vale a pena trocar o $1d20$ pelas outras alternativas, embora seja impossível a falha crítica (soma $1$) nos outros casos e o menor resultado possível para o $5d4$ seja $5$, diminuindo o espaço amostral apenas na faixa de valores que resultariam numa derrota no jogo de RPG."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "É possível calcular os valores exatos dessas probabilidades:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def brute_force_dice_sum_threshold_probability(quantity, sides, threshold=16, saturation=None):\n",
    "    return (sum(np.clip(np.arange(sides ** k)\n",
    "                          .repeat(sides ** (quantity - k)) % sides,\n",
    "                        0, (saturation or sides) - 1)\n",
    "                for k in range(1, quantity + 1)) >= threshold - quantity\n",
    "           ).mean()"
   ]
  },
  {
   "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>$1d20$</th>\n",
       "      <th>$2d10$</th>\n",
       "      <th>$5d4$</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Frequências relativas em que $X \\ge 16$</th>\n",
       "      <td>0.263</td>\n",
       "      <td>0.147</td>\n",
       "      <td>0.122000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>$P(X \\ge 16)$</th>\n",
       "      <td>0.250</td>\n",
       "      <td>0.150</td>\n",
       "      <td>0.118164</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                         $1d20$  $2d10$     $5d4$\n",
       "Frequências relativas em que $X \\ge 16$   0.263   0.147  0.122000\n",
       "$P(X \\ge 16)$                             0.250   0.150  0.118164"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "freqs_dframe.loc[\"$P(X \\ge 16)$\"] = {\n",
    "    \"$1d20$\": brute_force_dice_sum_threshold_probability(1, 20),\n",
    "    \"$2d10$\": brute_force_dice_sum_threshold_probability(2, 10),\n",
    "    \"$5d4$\": brute_force_dice_sum_threshold_probability(5, 4),\n",
    "}\n",
    "freqs_dframe"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Dado de 6 lados"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Se o dado tivesse $6$ lados, estas seriam as funções massa de probabilidade para a soma dos resultados de $n$ lançamentos (Imagens obtidas da Wikipedia):\n",
    "\n",
    "[![](Probability density function of the sum of two dice.svg)](https://en.wikipedia.org/wiki/File:Dice_Distribution_%28bar%29.svg)\n",
    "\n",
    "[![](Comparison of probability density functions for the sum of n dice to illustrate the central limit theorem.svg)](https://en.wikipedia.org/wiki/File:Dice_sum_central_limit_theorem.svg)\n",
    "\n",
    "A última imagem mostra as distribuições centralizadas e com um fator de escala ajustado de forma a ilustrar que essas somas aparentemente convergem para uma mesma distribuição."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Skewness (assimetria / obliquidade)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Depois de mostrar esses resultados para o mestre do RPG, já quase acusando ele de querer ludibriar os jogadores com um procedimento que diminui as chances de vitória, o próprio mestre fornece uma alternativa para se redimir: lançar $5d6$, mas \"saturar\" os valores dos dados $d6$ para que nunca ultrapassem $4$ individualmente. Dessa vez temos uma distribuição assimétrica!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXcAAAEKCAYAAADpfBXhAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n0P8H1/EpJHiP/qgAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f4ff4e2ca90>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.bar([1, 2, 3, 4], np.array([1, 1, 1, 3]) / 6)\n",
    "plt.title(\"Função massa de probabilidades do $d6$ saturado em $4$\");"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Efetuando o experimento $1000$ vezes, como realizado anteriormente, obtemos:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYsAAAEKCAYAAADjDHn2AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f4ff4fd4c50>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "events6 = np.random.randint(1, 7, size=5000)\n",
    "events6_clip4 = np.clip(events6, None, 4)\n",
    "plt.bar(range(5, 21), np.bincount(events6_clip4.reshape(-1, 5).sum(axis=1))[5:])\n",
    "plt.title(\"Histograma de $1000$ eventos $5d6$ com cada $d6$ saturado em $4$\");"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "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>$1d20$</th>\n",
       "      <th>$2d10$</th>\n",
       "      <th>$5d4$</th>\n",
       "      <th>$5d6$ saturado</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Frequências relativas em que $X \\ge 16$</th>\n",
       "      <td>0.263</td>\n",
       "      <td>0.147</td>\n",
       "      <td>0.122000</td>\n",
       "      <td>0.400000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>$P(X \\ge 16)$</th>\n",
       "      <td>0.250</td>\n",
       "      <td>0.150</td>\n",
       "      <td>0.118164</td>\n",
       "      <td>0.444059</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                         $1d20$  $2d10$     $5d4$  \\\n",
       "Frequências relativas em que $X \\ge 16$   0.263   0.147  0.122000   \n",
       "$P(X \\ge 16)$                             0.250   0.150  0.118164   \n",
       "\n",
       "                                         $5d6$ saturado  \n",
       "Frequências relativas em que $X \\ge 16$        0.400000  \n",
       "$P(X \\ge 16)$                                  0.444059  "
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "freqs_dframe[\"$5d6$ saturado\"] = [\n",
    "    (events6_clip4.reshape(-1, 5).sum(axis=1) >= 16).mean(),\n",
    "    brute_force_dice_sum_threshold_probability(5, 6, saturation=4),\n",
    "]\n",
    "freqs_dframe"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Dessa vez vale a pena aceitar a proposta do mestre! Porém, o que aconteceria com a distribuição se lançássemos $25d6$ com essa saturação em $4$ para cada dado?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYwAAAEKCAYAAAAB0GKPAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f4ff4d276a0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "events6_more = np.random.randint(1, 7, size=25000)\n",
    "events25d6_clip4 = np.clip(events6_more, None, 4).reshape(-1, 25).sum(axis=1)\n",
    "plt.hist(events25d6_clip4, bins=events25d6_clip4.ptp() + 1)\n",
    "plt.title(\"Histograma de $1000$ eventos $25d6$ com cada $d6$ saturado em $4$\");"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "O $d6$ saturado é basicamente um $d4$ viciado ($50\\%$ das vezes resulta em $4$), porém essa distribuição de $25d6$ saturado está quase simétrica!\n",
    "\n",
    "Embora a convergência na direção da distribuição normal não seja uniforme (as caudas da distribuição levam mais iterações que a região central), esse é um resultado prático previsto por aqueles que conhecem o Teorema do Limite Central: a distribuição da soma de $n$ variáveis aleatórias *i.i.d.* converge em distribuição para a normal com o aumento do $n$, mesmo que a distribuição original seja assimétrica."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Para uma ilustração interativa, veja: http://onlinestatbook.com/stat_sim/sampling_dist/index.html"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Distribuição normal"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "A distribuição normal/gaussiana $\\mathcal{N}(\\mu, \\sigma^2)$ é dada pela equação:\n",
    "\n",
    "$$\n",
    "\\displaystyle f(x | \\mu, \\sigma^2) =\n",
    "\\frac{\\displaystyle 1}\n",
    "     {\\displaystyle \\sqrt{2 \\pi \\sigma^2}}\n",
    "\\cdot\n",
    "e^{\\displaystyle - \\frac{\\textstyle(x - \\mu)^2}\n",
    "                        {\\textstyle2 \\sigma^2}}\n",
    "$$\n",
    "\n",
    "\n",
    "Essa é a função densidade de probabilidade para média $\\mu$ e variância $\\sigma^2$."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "A distribuição normal padrão é aquela com média zero e variância unitária, $\\mathcal{N}(0, 1)$:\n",
    "\n",
    "$$\n",
    "\\displaystyle \\phi(x) =\n",
    "\\frac{\\displaystyle 1}\n",
    "     {\\displaystyle \\sqrt{2 \\pi}}\n",
    "\\cdot\n",
    "e^{\\displaystyle - x^2 / 2}\n",
    "$$"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAX0AAAEKCAYAAAD+XoUoAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\ngg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f4ff5003630>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "xs = np.linspace(-3, 3, 200)\n",
    "plt.plot(xs, stats.norm.pdf(xs))\n",
    "plt.title(\"Função densidade de probabilidade da $\\mathcal{N}(0, 1)$\");"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Essa distribuição possui diversas propriedades importantes, por exemplo a soma de duas variáveis aleatórias com distribuição normal resulta em uma variável aleatória com distribuição normal."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Erro da média"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Se $\\bar{X}$ é a média de uma amostra, a distribuição da variáveis aleatórias $\\bar{X}$ é diferente da distribuição dos $X$ originais. As médias se mantém:"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "$$\\require{cancel}\n",
    "E[\\bar{X}]\n",
    "= E \\left[ \\frac{1}{n} \\sum_{i=1}^{n} X_i \\right]\n",
    "= \\frac{1}{n} \\sum_{i=1}^{n} E[X_i]\n",
    "= \\frac{1}{\\cancel{n}} \\cancel{ \\left[ \\sum_{i=1}^{n} 1 \\right] } E[X]\n",
    "= E[X]\n",
    "$$"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Por outro lado, a variância da distribuição das médias não é igual à variância da distribuição original:"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "$$\\require{cancel}\n",
    "\\operatorname{var} \\bar{X}\n",
    "= \\operatorname{var} \\left[ \\frac{1}{n} \\sum_{i=1}^{n} X_i \\right]\n",
    "= \\frac{1}{n^2} \\sum_{i=1}^{n} \\operatorname{var} X_i\n",
    "= \\frac{1}{n^\\cancel{2}} \\cancel{ \\left[ \\sum_{i=1}^{n} 1 \\right] } \\operatorname{var} X\n",
    "= \\frac{\\operatorname{var} X}{n}\n",
    "$$"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Isso é importante neste contexto: se o desvio padrão da distribuição original é $\\sigma$, o erro padrão da média é $\\sigma \\left/ \\sqrt{n} \\right.$."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Formalização do CLT (Central Limit Theorem)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Há mais de uma forma de provar o teorema do limite central. O que segue é uma prova fundamentada na função característica, utilizando, ao final, o teorema da convergência de Lévy."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "A formulação clássica pode ser ilustrada por esta imagem (obtida da Wikipedia), mas vamos adotar uma conversão para evitar os deslocamentos de média e desvio padrão:\n",
    "\n",
    "[![](IllustrationCentralTheorem.png)](https://en.wikipedia.org/wiki/File:IllustrationCentralTheorem.png)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Seja ${X_i}$ uma sequência de variáveis aleatórias *i.i.d.* com média $E[X_i] = \\mu$ e variância $\\operatorname{var}(X_i) = \\sigma^2 > 0$. A soma dos $n$ primeiros elementos da sequência é:\n",
    "\n",
    "$$\n",
    "S_n = \\sum_{i=1}^n X_i\n",
    "$$\n",
    "\n",
    "Queremos mostrar que:\n",
    "\n",
    "$$\n",
    "\\bar{X}\n",
    "\\xrightarrow{\\mathcal{D}}\n",
    "\\mathcal{N} \\left( \\mu, \\frac{\\sigma^2}{n} \\right)\n",
    "$$\n",
    "\n",
    "Ou,\n",
    "\n",
    "$$\n",
    "S_n\n",
    "\\xrightarrow{\\mathcal{D}}\n",
    "\\mathcal{N} \\left( n \\mu, n \\sigma^2 \\right)\n",
    "$$\n",
    "\n",
    "Ou, equivalentemente,\n",
    "\n",
    "$$\n",
    "Z_n\n",
    "=\n",
    "\\frac{\\displaystyle \\frac{1}{n} \\sum_{i=1}^n \\left( X_i - \\mu \\right)}\n",
    "     {\\sigma \\left/ \\sqrt{n} \\right.}\n",
    "=\n",
    "\\frac{S_n - n \\mu}{\\sqrt{n \\sigma^2}}\n",
    "\\xrightarrow{\\mathcal{D}}\n",
    "\\mathcal{N}(0, 1)\n",
    "$$\n",
    "\n",
    "Em que o símbolo $\\xrightarrow{\\mathcal{D}}$ denota que o lado esquerdo converge em distribuição ao lado direito para valores maiores de $n$."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Função característica"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "A *função característica* de uma distribuição é o valor esperado da exponencial complexa $e^{i t X}$, consistindo fundamentalmente em uma inversa da transformada de Fourier da função densidade de probabilidade.\n",
    "\n",
    "$$\n",
    "cf_X(t) = E[e^{i t X}] = E[exp(i t X)]\n",
    "$$"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "No caso da distribuição normal, a função característica é:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "exp(t*(I*mu - sigma**2*t/2))"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X = Normal(\"X\", mu, sigma)\n",
    "E(exp(I * t * X)).simplify() # Characteristic Function da N(mu, sigma^2)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Ou:\n",
    "\n",
    "$$\n",
    "cf_{\\mathcal{N}(\\mu, \\sigma^2)}(t)\n",
    "=\n",
    "\\exp \\left( i \\mu t - \\frac{1}{2} \\sigma^2 t^2 \\right)\n",
    "$$\n",
    "\n",
    "Para a $\\mathcal{N}(0, 1)$, temos:\n",
    "\n",
    "$$\n",
    "cf_{\\mathcal{N}(0, 1)}(t)\n",
    "=\n",
    "\\exp \\left( - \\frac{t^2}{2} \\right)\n",
    "$$"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "exp(-t**2/2)"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cf = E(exp(I * t * X.subs({mu: 0, sigma: 1}))).simplify()\n",
    "cf # Characteristic Function da N(0, 1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Relação entre funções características"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Seja:\n",
    "\n",
    "$$\n",
    "Y_i = \\frac{X_i - \\mu}{\\sigma}\n",
    "$$\n",
    "\n",
    "Dessa forma, temos $E[Y_i] = 0$, por conta da subtração no numerador, e $\\operatorname{var}[Y_i] = 1$, por conta da normalização realizada pelo denominador."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "A anteriormente definida $Z_n$ pode então ser escrita como função de $Y_i$:\n",
    "\n",
    "$$\n",
    "Z_n\n",
    "=\n",
    "\\frac{\\displaystyle \\frac{1}{n} \\sum_{i=1}^n \\left( X_i - \\mu \\right)}\n",
    "     {\\sigma \\left/ \\sqrt{n} \\right.}\n",
    "=\n",
    "\\frac{1}{\\sqrt{n}}\n",
    "\\sum_{i=1}^n \\frac{X_i - \\mu}{\\sigma}\n",
    "=\n",
    "\\frac{1}{\\sqrt{n}}\n",
    "\\sum_{i=1}^n Y_i\n",
    "$$"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Calculando a função característica da $Z_n$, obtemos:\n",
    "\n",
    "$$\n",
    "\\begin{array}{rcl} \\displaystyle\n",
    "cf_{Z_n}(t)\n",
    "   & \\displaystyle = & \\displaystyle\n",
    "E\\left[ \\exp \\left( i t Z_n \\right) \\right]\n",
    "\\\\ & \\displaystyle = & \\displaystyle\n",
    "E\\left[ \\exp \\left( \\frac{i t}{\\sqrt{n}} \\sum_{j=1}^n Y_j \\right) \\right]\n",
    "\\\\ & \\displaystyle = & \\displaystyle\n",
    "E\\left[ \\prod_{j=1}^n \\exp \\left( \\frac{i t}{\\sqrt{n}} Y_j \\right) \\right]\n",
    "\\\\ & \\displaystyle \\stackrel{i.i.d.}{=} & \\displaystyle\n",
    "\\prod_{j=1}^n E\\left[ \\exp \\left( i \\frac{t}{\\sqrt{n}} Y_j \\right) \\right]\n",
    "\\\\ & \\displaystyle = & \\displaystyle\n",
    "\\prod_{j=1}^n cf_{Y_j}\\left( \\frac{t}{\\sqrt{n}} \\right)\n",
    "\\\\ & \\displaystyle = & \\displaystyle\n",
    "\\left[ cf_{Y}\\left( \\frac{t}{\\sqrt{n}} \\right) \\right]^n\n",
    "\\end{array}\n",
    "$$"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Expansão em série Taylor"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "A partir da série de Taylor / McLaurin de $e^{x}$, temos:\n",
    "\n",
    "$$\n",
    "\\sum_{k=0}^\\infty \\frac{x^k}{k!}\n",
    "$$"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Embora não estejamos utilizando, a função geradora de momentos é o valor esperado dessa última exponencial para $x = t X$, de forma que:\n",
    "\n",
    "$$\n",
    "M_X(t)\n",
    "=\n",
    "E\\left[ e^{t X} \\right]\n",
    "=\n",
    "E\\left[ \\sum_{k=0}^\\infty \\frac{(t X)^k}{k!} \\right]\n",
    "=\n",
    "\\sum_{k=0}^\\infty E\\left[ \\frac{X^k}{k!} t^k \\right]\n",
    "=\n",
    "\\sum_{k=0}^\\infty \\frac{E[X^k]}{k!} t^k\n",
    "$$"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Na função característica, a exponencial é complexa, nosso expoente é $x = i t X$:\n",
    "\n",
    "$$\n",
    "cf_X(t)\n",
    "=\n",
    "E\\left[ e^{i t X} \\right]\n",
    "=\n",
    "\\sum_{k=0}^\\infty \\frac{E[X^k]}{k!} t^k i^k\n",
    "$$"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Ou,\n",
    "\n",
    "$$\n",
    "cf_X(t)\n",
    "=\n",
    "1\n",
    "+ E(X) t i\n",
    "- \\frac{E(X^2)}{2} t^2\n",
    "- \\frac{E(X^3)}{3!} t^3 i\n",
    "+ \\frac{E(X^4)}{4!} t^4\n",
    "+ \\sum_{k=5}^\\infty \\frac{E[X^k]}{k!} t^k i^k\n",
    "$$"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Substituindo $X$ por $Y$, e $t$ por $t \\left/ \\sqrt{n} \\right.$,\n",
    "\n",
    "$$\n",
    "cf_Y\\left( \\frac{t}{\\sqrt{n}} \\right)\n",
    "=\n",
    "1\n",
    "+ \\frac{E(Y)}{\\sqrt{n}} t i\n",
    "- \\frac{E(Y^2)}{2 n} t^2\n",
    "- \\frac{E(Y^3)}{3! n \\sqrt{n}} t^3 i\n",
    "+ \\frac{E(Y^4)}{4! n^2} t^4\n",
    "+ \\sum_{k=5}^\\infty \\frac{E[Y^k]}{k! \\sqrt{n^k}} t^k i^k\n",
    "$$"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Para o caso da $cf_Y(\\cdot)$, nós sabemos que $E[Y] = 0$ e que $\\require{cancel} \\operatorname{var}(Y) = E[Y^2] - \\cancelto{0}{E[Y]^2} = 1$, logo:\n",
    "\n",
    "$$\\require{cancel}\n",
    "cf_Y\\left( \\frac{t}{\\sqrt{n}} \\right)\n",
    "=\n",
    "E\\left[ \\exp \\left( i \\frac{t}{\\sqrt{n}} Y \\right) \\right]\n",
    "=\n",
    "1\n",
    "+ \\cancel{0 \\frac{t}{\\sqrt{n}} i}\n",
    "- \\frac{1}{2} \\frac{t^2}{n}\n",
    "- \\frac{E(Y^3)}{3!} \\frac{t^3}{n \\sqrt{n}} i\n",
    "+ \\frac{E(Y^4)}{4!} \\frac{t^4}{n^2}\n",
    "+ \\sum_{k=5}^\\infty \\frac{E[Y^k]}{k! \\sqrt{n^k}} t^k i^k\n",
    "$$"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Deixando fora do somatório somente os termos sobre os quais temos a informação completa,\n",
    "\n",
    "$$\n",
    "cf_Y\\left( \\frac{t}{\\sqrt{n}} \\right)\n",
    "=\n",
    "1 - \\frac{t^2}{2 n} + \\sum_{k=3}^\\infty \\frac{E[Y^k]}{k! \\sqrt{n^k}} t^k i^k\n",
    "$$"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Esse resultado vale para qualquer variável $Y$ definida como *z-score* de outra distribuição, como fizemos com $X$. Em particular, o polinômio de Taylor de segunda ordem da $\\mathcal{N}(0, 1)$ também possui essa configuração (exemplo com $n = 1$):"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "-t**2/2 + 1"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cf.subs({t: 0}) + t * cf.diff(t).subs({t: 0}) + t ** 2 * cf.diff(t, 2).subs({t: 0}) / 2"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Finalizando a prova do teorema"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Queremos saber o que acontece com a função característica de $Z_n$ quando $n$ é grande:\n",
    "\n",
    "$$\n",
    "\\begin{array}{rcl} \\displaystyle\n",
    "\\lim_{n \\to \\infty} cf_{Z_n}(t)\n",
    "   & \\displaystyle = & \\displaystyle\n",
    "\\lim_{n \\to \\infty} \\left[ cf_{Y}\\left( \\frac{t}{\\sqrt{n}} \\right) \\right]^n\n",
    "\\\\ & \\displaystyle = & \\displaystyle\n",
    "\\lim_{n \\to \\infty} \\left[ 1 - \\frac{t^2}{2 n} + \\sum_{k=3}^\\infty \\frac{E[Y^k]}{k! \\sqrt{n^k}} t^k i^k \\right]^n\n",
    "\\end{array}\n",
    "$$"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Ao invés disso, se tentássemos obter o limite do logaritmo natural da função característica de $Z_n$, teríamos:\n",
    "\n",
    "$$\n",
    "\\begin{array}{rcl} \\displaystyle\n",
    "\\lim_{n \\to \\infty} \\ln \\left[ cf_{Z_n}(t) \\right]\n",
    "   & \\displaystyle = & \\displaystyle\n",
    "\\lim_{n \\to \\infty} \\ln \\left[ 1 - \\frac{t^2}{2 n} + \\sum_{k=3}^\\infty \\frac{E[Y^k]}{k! \\sqrt{n^k}} t^k i^k \\right]^n\n",
    "\\\\ & \\displaystyle = & \\displaystyle\n",
    "\\lim_{n \\to \\infty} n \\; \\ln \\left[ 1 - \\frac{t^2}{2 n} + \\sum_{k=3}^\\infty \\frac{E[Y^k]}{k! \\sqrt{n^k}} t^k i^k \\right]\n",
    "\\\\ & \\displaystyle \\stackrel{n = 1 / m}{=} & \\displaystyle\n",
    "\\lim_{m \\to 0}\n",
    "  \\frac{\\displaystyle \\ln \\left[ 1 - \\frac{t^2 m}{2} + \\sum_{k=3}^\\infty \\frac{E[Y^k]}{k!} t^k i^k \\sqrt{m^k} \\right]}\n",
    "       {\\displaystyle m}\n",
    "\\\\ & \\displaystyle \\stackrel{L'Hôpital}{=} & \\displaystyle\n",
    "\\lim_{m \\to 0}\n",
    "  \\frac{\\displaystyle - \\frac{t^2}{2} + \\sum_{k=3}^\\infty \\frac{k}{2} \\frac{E[Y^k]}{k!} t^k i^k m^{k/2 - 1}}\n",
    "       {\\displaystyle 1 - \\frac{t^2 m}{2} + \\sum_{k=3}^\\infty \\frac{E[Y^k]}{k!} t^k i^k \\sqrt{m^k}}\n",
    "\\\\ & \\displaystyle = & \\displaystyle\n",
    "\\lim_{m \\to 0} \\frac{- \\frac{t^2}{2}\n",
    "                     + \\cancel{\\frac{3}{2} \\frac{E[Y^3]}{3!} t^3 i^3 \\sqrt{m}}\n",
    "                     + \\cancel{\\frac{4}{2} \\frac{E[Y^4]}{4!} t^4 i^4 m}\n",
    "                     + \\cancel{\\frac{5}{2} \\frac{E[Y^5]}{5!} t^5 i^5 m \\sqrt{m}}\n",
    "                     + \\cdots\n",
    "               }{1 - \\cancel{\\frac{t^2 m}{2}}\n",
    "                   + \\cancel{\\frac{E[Y^3]}{3!} t^3 i^3 m \\sqrt{m}}\n",
    "                   + \\cancel{\\frac{E[Y^4]}{4!} t^4 i^4 m^2}\n",
    "                   + \\cancel{\\frac{E[Y^5]}{5!} t^5 i^5 m^2 \\sqrt{m}}\n",
    "                   + \\cdots\n",
    "               }\n",
    "\\\\ & \\displaystyle = & \\displaystyle\n",
    "- \\frac{t^2}{2}\n",
    "\\end{array}\n",
    "$$\n",
    "\n",
    "Onde usamos a regra de L'Hôpital pelo fato de que tanto o numerador como o denominador do limite convergiam para zero."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Logo,\n",
    "\n",
    "$$\n",
    "\\lim_{n \\to \\infty} cf_{Z_n}(t)\n",
    "=\n",
    "\\exp\\left\\{ \\lim_{n \\to \\infty} \\ln \\left[ cf_{Z_n}(t) \\right] \\right\\}\n",
    "=\n",
    "e^{\\textstyle -t^2 / 2}\n",
    "$$"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "E essa exponencial é exatamente a função característica da $\\mathcal{N}(0, 1)$. Pelo teorema da convergência de Lévy, sabemos que a convergência ponto-a-ponto da função característica nos leva à convergência em distribuição.\n",
    "\n",
    "**Q.E.D.**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Fim!"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Resumidamente, o estudo da distribuição das somas e da distribuição das médias de $n$ elementos de outra distribuição pode ser aproximado pelo estudo da distribuição normal, quando $n$ é grande."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Espero que isto tenha sido útil! =)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}