├── README.md ├── banner.png.png ├── cap1_tiposvariaveis_escala_mensuracao_precisao.ipynb ├── cap2_estatistica_descritiva_univariada.ipynb ├── cap3_estatistica_descritiva_bivariada.ipynb ├── cap4_introducao_probabilidade.ipynb ├── cap5_variavel_aleatori_dist_probabilidade.ipynb ├── cap6_estatistica_inferencial.ipynb └── cap7_teste_hipotese.ipynb /README.md: -------------------------------------------------------------------------------- 1 | # Manual de Análise de Dados - Python 2 | 3 | ![Project Image](banner.png.png) 4 | 5 | ## Sobre o Projeto 6 | 7 | Bem-vindo ao Manual de Análise de Dados em Python! Neste repositório, você encontrará exercícios e exemplos práticos adaptados do livro "Manual de Análise de Dados Estatística e Modelagem Multivariada com Excel, SPSS, Stata" pelos professores Luiz Paulo Fávero e Patrícia Belfiore. Aqui, você terá a oportunidade de explorar conceitos importantes de análise de dados utilizando a linguagem de programação Python. 8 | 9 | Este repositório é um recurso de estudo valioso que complementa um dos livros mais renomados na área de análise de dados. Esperamos que ele o ajude a aprimorar suas habilidades e conhecimentos em análise de dados. 10 | 11 | ## Tecnologias Utilizadas 12 | 13 | - Python 3.10.11 14 | - Pandas 1.5.3 15 | - Numpy 1.22.4 16 | - Matplotlib 3.7.1 17 | - Seaborn 0.12.2 18 | - Scipy 1.10.1 19 | - Scikit-learn 1.2.2 20 | 21 | ## Capítulos 22 | - [`cap1_tiposvariaveis_escala_mensuracao_precisao`](https://github.com/jvataidee/ManualdeAnalisedeDadosPython/blob/master/cap1_tiposvariaveis_escala_mensuracao_precisao.ipynb): Neste capítulo, são abordados os diferentes tipos de variáveis, escalas de mensuração e a importância da precisão nos dados estatísticos. 23 | - [`cap2_estatistica_descritiva_univariada`](https://github.com/jvataidee/ManualdeAnalisedeDadosPython/blob/master/cap2_estatistica_descritiva_univariada.ipynb):O segundo capítulo explora a estatística descritiva univariada, que envolve a análise e interpretação de dados de uma única variável. 24 | - [`cap3_estatistica_descritiva_bivariada`](https://github.com/jvataidee/ManualdeAnalisedeDadosPython/blob/master/cap3_estatistica_descritiva_bivariada.ipynb):Neste capítulo, a estatística descritiva bivariada é apresentada, fornecendo métodos para analisar a relação entre duas variáveis e extrair insights relevantes. 25 | - [`cap4_introducao_probabilidade`](https://github.com/jvataidee/ManualdeAnalisedeDadosPython/blob/master/cap4_introducao_probabilidade.ipynb):O capítulo quatro introduz os conceitos básicos de probabilidade, permitindo uma compreensão mais profunda da incerteza e dos eventos aleatórios. 26 | - [`cap5_variavel_aleatori_dist_probabilidade`](https://github.com/jvataidee/ManualdeAnalisedeDadosPython/blob/master/cap5_variavel_aleatori_dist_probabilidade.ipynb):A variável aleatória e a distribuição de probabilidade são discutidas neste capítulo, fornecendo ferramentas para modelar e analisar eventos aleatórios. 27 | - [`cap6_estatistica_inferencial`](https://github.com/jvataidee/ManualdeAnalisedeDadosPython/blob/master/cap6_estatistica_inferencial.ipynb):Neste capítulo, é abordada a estatística inferencial, que permite fazer inferências e generalizações sobre uma população com base em uma amostra. 28 | - [`cap7_teste_hipotese`](https://github.com/jvataidee/ManualdeAnalisedeDadosPython/blob/master/cap7_teste_hipotese.ipynb):O capítulo sete trata dos testes de hipóteses em, onde são aplicados métodos estatísticos para avaliar e tomar decisões sobre diferentes cenários ou tratamentos. *em produção ...* 29 | 30 | 31 | ## Como Começar 32 | Para começar a explorar o Manual de Análise de Dados em Python, siga estas etapas: 33 | 34 | 1. Clone este repositório para o seu ambiente local. 35 | 2 .Instale as dependências necessárias, como Python, Pandas, Numpy, Matplotlib, etc. 36 | 3. Navegue pelos notebooks Jupyter disponíveis para cada capítulo do livro. Cada capítulo contém exemplos práticos e exercícios para você praticar e aprofundar seu conhecimento em análise de dados com Python. 37 | 38 | ## Recursos 39 | 40 | O livro "Manual de Análise de Dados" é altamente recomendado para acompanhar este repositório. Aqui está o link para compra: [Manual de Análise de Dados](https://www.amazon.com.br/Manual-An%C3%A1lise-Dados-Luiz-F%C3%A1vero/dp/8535270876) ou a edição digital [Grupo Gen](https://www.grupogen.com.br/) 41 | 42 | ## Contato 43 | 44 | - [Blog](www.joaoataide.com) 45 | - [LinkedIn](https://www.linkedin.com/in/joaoataidee/) 46 | - Email: contato@joaoataide.com 47 | -------------------------------------------------------------------------------- /banner.png.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/jvataidee/ManualdeAnalisedeDadosPython/3bf6f7f40c21a8db083f8911c091efcc0a77a3ca/banner.png.png -------------------------------------------------------------------------------- /cap4_introducao_probabilidade.ipynb: -------------------------------------------------------------------------------- 1 | { 2 | "nbformat": 4, 3 | "nbformat_minor": 0, 4 | "metadata": { 5 | "colab": { 6 | "name": "cap4_introducao_probabilidade.ipynb", 7 | "provenance": [], 8 | "collapsed_sections": [], 9 | "toc_visible": true, 10 | "include_colab_link": true 11 | }, 12 | "kernelspec": { 13 | "name": "python3", 14 | "display_name": "Python 3" 15 | }, 16 | "accelerator": "GPU" 17 | }, 18 | "cells": [ 19 | { 20 | "cell_type": "markdown", 21 | "metadata": { 22 | "id": "view-in-github", 23 | "colab_type": "text" 24 | }, 25 | "source": [ 26 | "\"Open" 27 | ] 28 | }, 29 | { 30 | "cell_type": "markdown", 31 | "metadata": { 32 | "id": "kLI0gJtzDGuQ" 33 | }, 34 | "source": [ 35 | "**by: [João Ataíde](https://www.joaoataide.com)**\n", 36 | "# **CAP 04: Introdução a Probabilidade**\n", 37 | "\n", 38 | "---\n", 39 | "\n" 40 | ] 41 | }, 42 | { 43 | "cell_type": "code", 44 | "source": [ 45 | "#importar bibliotecas\n", 46 | "import pandas as pd\n", 47 | "import numpy as np\n", 48 | "import math\n", 49 | "import matplotlib.pyplot as plt\n", 50 | "import seaborn as sns\n", 51 | "import statistics as st\n", 52 | "import scipy.stats as stats\n", 53 | "from scipy.stats import chi2_contingency, spearmanr" 54 | ], 55 | "metadata": { 56 | "id": "dTSRYPl7ORJW" 57 | }, 58 | "execution_count": 30, 59 | "outputs": [] 60 | }, 61 | { 62 | "cell_type": "markdown", 63 | "source": [ 64 | "## **Conceitos**\n", 65 | "\n", 66 | "### **Experimento aleatório**\n", 67 | "Fenômeno imprevisível, se repetido o experimento não é possível aparecer o mesmo resultado\n", 68 | "\n", 69 | "### **Espaço amostral**\n", 70 | "Consiste $S$ em todos o resultados possíveis do experimento\n", 71 | "\n" 72 | ], 73 | "metadata": { 74 | "id": "zTllLwnt6IGI" 75 | } 76 | }, 77 | { 78 | "cell_type": "code", 79 | "source": [ 80 | "S = list(range(60))\n", 81 | "len(S)" 82 | ], 83 | "metadata": { 84 | "colab": { 85 | "base_uri": "https://localhost:8080/" 86 | }, 87 | "id": "4fATXMdFe8KZ", 88 | "outputId": "a1b23640-aeb9-4761-f0e9-c68ae00cb998" 89 | }, 90 | "execution_count": 31, 91 | "outputs": [ 92 | { 93 | "output_type": "execute_result", 94 | "data": { 95 | "text/plain": [ 96 | "60" 97 | ] 98 | }, 99 | "metadata": {}, 100 | "execution_count": 31 101 | } 102 | ] 103 | }, 104 | { 105 | "cell_type": "markdown", 106 | "source": [ 107 | "\n", 108 | "### **Eventos**\n", 109 | "Subconjunto de um espaço amostral, uma fenômeno que ocorre dentro de um conjunto de resultados possíveis\n" 110 | ], 111 | "metadata": { 112 | "id": "_fUNB87pe3y8" 113 | } 114 | }, 115 | { 116 | "cell_type": "markdown", 117 | "source": [ 118 | "\n", 119 | "### **Conjuntos**\n", 120 | "\n", 121 | "* `União`\n", 122 | "Eventos $A$ ∪ $B$, gera no um novo evento resultante de todas os resultados dos dois eventos.\n", 123 | "\n", 124 | "* `Intersecção`\n", 125 | "Eventos $A$ ∩ $B$, gera um novo evento resultante da diferença entre os dois evento, sendo que esse tem que percencer aos dois.\n", 126 | "\n", 127 | "* `Complementares`\n", 128 | "Evento $C$ é um evento que contês todos os elementos do espaço amostral.\n", 129 | "\n" 130 | ], 131 | "metadata": { 132 | "id": "QwN2vP1_e5Ko" 133 | } 134 | }, 135 | { 136 | "cell_type": "markdown", 137 | "source": [ 138 | "### **Eventos independentes**\n", 139 | "Evento $B$ não impacta na condição da ocrrência do evento $A$\n", 140 | "\n" 141 | ], 142 | "metadata": { 143 | "id": "hGFYmXBKe6DI" 144 | } 145 | }, 146 | { 147 | "cell_type": "markdown", 148 | "source": [ 149 | "### **Evento mutualmente excludente**\n", 150 | "O evento $A$ e o evento $B$ não possuem nenhum elemento em comum, não podendo ocorrer simutanemamento." 151 | ], 152 | "metadata": { 153 | "id": "MX1mj-0ze68H" 154 | } 155 | }, 156 | { 157 | "cell_type": "markdown", 158 | "source": [ 159 | "## **Definição de probabilidade**\n", 160 | "É conceito matemático que determina a ocorrência de uma determinado evento $A$ dentro de um espaço amostral\n", 161 | "\n", 162 | "$P(A) = \\frac{n_{A}}{n}$" 163 | ], 164 | "metadata": { 165 | "id": "Yyj2ZkF_cA0x" 166 | } 167 | }, 168 | { 169 | "cell_type": "code", 170 | "source": [ 171 | "dados = [1, 2, 3, 4, 5, 6]\n", 172 | "A = [2, 4, 6]\n", 173 | "\n", 174 | "probabilidade = len(A)/len(dados) * 100\n", 175 | "probabilidade" 176 | ], 177 | "metadata": { 178 | "colab": { 179 | "base_uri": "https://localhost:8080/" 180 | }, 181 | "id": "_x9mfpuvcdcL", 182 | "outputId": "7140c6e1-7542-4556-f4aa-db76297d5240" 183 | }, 184 | "execution_count": 32, 185 | "outputs": [ 186 | { 187 | "output_type": "execute_result", 188 | "data": { 189 | "text/plain": [ 190 | "50.0" 191 | ] 192 | }, 193 | "metadata": {}, 194 | "execution_count": 32 195 | } 196 | ] 197 | }, 198 | { 199 | "cell_type": "markdown", 200 | "source": [ 201 | "## **Regras básicas**\n", 202 | " \n", 203 | "\n", 204 | "1. `Variação probabilidade:` probabilidade sempre entre 0% a 100% ou $0-1$\n", 205 | "2. `Probabilidade espaço amostral:` probabilidade $S$ sempre é igual 1\n", 206 | "3. `Conjunto vazio:` se não tem dado não tem probabilidade\n", 207 | "4. `Adição de probabilidade: ` $P(A ∪ B) = P(A) + P(B) - P(A ∩ B)$\n", 208 | "5. `Probabilidade eventos complementares:` $P(A^c) = 1 - P(A)$\n", 209 | "6. `Multiplicação de probabilidade: ` $P(A ∩ B) = P(A)P(B)$\n", 210 | "\n" 211 | ], 212 | "metadata": { 213 | "id": "jPhKTlfEc5OV" 214 | } 215 | }, 216 | { 217 | "cell_type": "markdown", 218 | "source": [ 219 | "## **Probabilidade Condicional**\n", 220 | "Quando os eventos não forem independente deve usar conceito de probabilidade condicional, basicamente se tive um evento qual a probabilidade de acontecer o outro.\n", 221 | "\n", 222 | "$P(A|B) = \\frac{P(A ∩ B)}{P(B)}$\n", 223 | "\n", 224 | "* `multiplicacao probabilidade: ` $P(A ∩ B) = P(A)P(B|A)$ pode ser extendido para o terceiro evento multiplicando por $P(C|A ∩ B)$" 225 | ], 226 | "metadata": { 227 | "id": "lapWJ-DVGdJ1" 228 | } 229 | }, 230 | { 231 | "cell_type": "markdown", 232 | "source": [ 233 | "## **Teorema de Bayes**\n", 234 | "Uma probabilidade calculada de um evento e irá ser colocado mais informações, esse teorema possibilita recalcular.\n", 235 | "\n", 236 | "$P(B_{i} | A) = \\frac{P(B_{i} ∩ A)}{P(A)} $" 237 | ], 238 | "metadata": { 239 | "id": "i0qklLosN6Dv" 240 | } 241 | }, 242 | { 243 | "cell_type": "markdown", 244 | "source": [ 245 | "## **Análise Combinatória**" 246 | ], 247 | "metadata": { 248 | "id": "FlvddR48XNcN" 249 | } 250 | }, 251 | { 252 | "cell_type": "markdown", 253 | "source": [ 254 | "## **Probpy v.0.0**\n", 255 | "by: João Ataíde" 256 | ], 257 | "metadata": { 258 | "id": "5ryrKKeLGALP" 259 | } 260 | }, 261 | { 262 | "cell_type": "code", 263 | "source": [ 264 | "class universo:\n", 265 | " def __init__(self, eventoA, espaco_amostral, eventoB = None,\n", 266 | " eventoC = None, novoseventos = None):\n", 267 | " self.eventoA = eventoA\n", 268 | " self.eventoB = eventoB\n", 269 | " self.eventoC = eventoC\n", 270 | " self.espaco_amostral= espaco_amostral\n", 271 | "\n", 272 | " def prob(self):\n", 273 | " return len(self.eventoA) / len(self.espaco_amostral)\n", 274 | "\n", 275 | " def intercet(self):\n", 276 | " return list(set(self.eventoA).intersection(self.eventoB))\n", 277 | " \n", 278 | " def union(self):\n", 279 | " return self.eventoA + self.eventoB\n", 280 | "\n", 281 | " def adicao_multual(self):\n", 282 | " prob_a = universo(self.eventoA, self.espaco_amostral).prob() \n", 283 | " prob_b = universo(self.eventoB, self.espaco_amostral).prob()\n", 284 | " intecect = universo(universo(self.eventoA,self.espaco_amostral, self.eventoB).intercet(), S).prob()\n", 285 | " return prob_a + prob_b - intecect\n", 286 | "\n", 287 | " def adicao_excludantes(self):\n", 288 | " prob_a = universo(self.eventoA, self.espaco_amostral).prob() \n", 289 | " prob_b = universo(self.eventoB, self.espaco_amostral).prob()\n", 290 | " return prob_a + prob_b\n", 291 | " \n", 292 | " def complement(self):\n", 293 | " return (1 - universo(self.eventoA, self.espaco_amostral).prob())\n", 294 | "\n", 295 | " def indepedent(self):\n", 296 | " return universo(self.eventoA, self.espaco_amostral).prob() * universo(self.eventoB, self.espaco_amostral).prob()\n", 297 | "\n", 298 | " def conditional(self):\n", 299 | " intecect = universo(universo(self.eventoA, self.espaco_amostral,\n", 300 | " self.eventoB).intercet(), self.espaco_amostral).prob()\n", 301 | " prob_B = universo(self.eventoB, self.espaco_amostral).prob()\n", 302 | " return intecect / prob_B\n", 303 | "\n", 304 | " def conditional_mult(self):\n", 305 | " interc = universo(self.eventoA, self.espaco_amostral, self.eventoB).intercet()\n", 306 | " intecect = len(interc) / len(self.espaco_amostral)\n", 307 | " prob_A = len(self.eventoA) / len(self.espaco_amostral)\n", 308 | " prob_B = len(self.eventoB) / len(self.espaco_amostral)\n", 309 | " prob_C = len(self.eventoC) / len(self.espaco_amostral)\n", 310 | " return prob_A * (intecect / prob_B) * (intecect / prob_C)\n" 311 | ], 312 | "metadata": { 313 | "id": "qqxXfZfjoE6u" 314 | }, 315 | "execution_count": 33, 316 | "outputs": [] 317 | }, 318 | { 319 | "cell_type": "markdown", 320 | "source": [ 321 | "## **Exercícios Probabilidade Básica**\n", 322 | "\n", 323 | "\n" 324 | ], 325 | "metadata": { 326 | "id": "1yQgxiFXfmWA" 327 | } 328 | }, 329 | { 330 | "cell_type": "markdown", 331 | "source": [ 332 | "### Exemplo 1" 333 | ], 334 | "metadata": { 335 | "id": "yRagkP10I3Yg" 336 | } 337 | }, 338 | { 339 | "cell_type": "code", 340 | "source": [ 341 | "#impar\n", 342 | "def impar(lista):\n", 343 | " return list(filter(lambda x: (x%2 != 0) , lista))\n", 344 | "\n", 345 | "\n", 346 | "#numeros par\n", 347 | "def par(lista):\n", 348 | " return list(filter(lambda x: (x%2 == 0) , lista))\n", 349 | "\n", 350 | "\n", 351 | "#multiplo de 5\n", 352 | "def mult_cinco(lista):\n", 353 | " return list(filter(lambda x: (x%5 == 0) , lista))\n", 354 | "\n", 355 | "\n", 356 | "#multiplo de 5\n", 357 | "def nao_mult_cinco(lista):\n", 358 | " return list(filter(lambda x: (x%5 != 0) , lista))" 359 | ], 360 | "metadata": { 361 | "id": "SuzMz0YSe0wC" 362 | }, 363 | "execution_count": 34, 364 | "outputs": [] 365 | }, 366 | { 367 | "cell_type": "markdown", 368 | "source": [ 369 | "### Exemplo 2" 370 | ], 371 | "metadata": { 372 | "id": "HSr68WUhI51G" 373 | } 374 | }, 375 | { 376 | "cell_type": "code", 377 | "source": [ 378 | "A = impar(S)\n", 379 | "B = mult_cinco(S)" 380 | ], 381 | "metadata": { 382 | "id": "yqlAfVXcuZAP" 383 | }, 384 | "execution_count": 35, 385 | "outputs": [] 386 | }, 387 | { 388 | "cell_type": "code", 389 | "source": [ 390 | "#B) Probabildiade evento inpar em S\n", 391 | "prob_A = universo(A, S)\n", 392 | "prob_A.prob()" 393 | ], 394 | "metadata": { 395 | "colab": { 396 | "base_uri": "https://localhost:8080/" 397 | }, 398 | "id": "rOkOlvLpfrSF", 399 | "outputId": "075d51b5-739c-497d-afca-6625219003dc" 400 | }, 401 | "execution_count": 36, 402 | "outputs": [ 403 | { 404 | "output_type": "execute_result", 405 | "data": { 406 | "text/plain": [ 407 | "0.5" 408 | ] 409 | }, 410 | "metadata": {}, 411 | "execution_count": 36 412 | } 413 | ] 414 | }, 415 | { 416 | "cell_type": "markdown", 417 | "source": [ 418 | "### Exemplo 3" 419 | ], 420 | "metadata": { 421 | "id": "kKLYrWfJI71v" 422 | } 423 | }, 424 | { 425 | "cell_type": "code", 426 | "source": [ 427 | "#C) Probabilidade evento multiplo de 5 em S\n", 428 | "prob_B = universo(B, S)\n", 429 | "prob_B.prob()" 430 | ], 431 | "metadata": { 432 | "colab": { 433 | "base_uri": "https://localhost:8080/" 434 | }, 435 | "id": "yy3op-oJiEvs", 436 | "outputId": "26bd3a5f-706c-42e5-aea6-3fac5108d5c5" 437 | }, 438 | "execution_count": 37, 439 | "outputs": [ 440 | { 441 | "output_type": "execute_result", 442 | "data": { 443 | "text/plain": [ 444 | "0.2" 445 | ] 446 | }, 447 | "metadata": {}, 448 | "execution_count": 37 449 | } 450 | ] 451 | }, 452 | { 453 | "cell_type": "markdown", 454 | "source": [ 455 | "### Exemplo 4" 456 | ], 457 | "metadata": { 458 | "id": "OYVvIXvFI9tD" 459 | } 460 | }, 461 | { 462 | "cell_type": "code", 463 | "source": [ 464 | "#Universo de estudo\n", 465 | "univers = universo(A, S, B)" 466 | ], 467 | "metadata": { 468 | "id": "qw-b5M1vgvSl" 469 | }, 470 | "execution_count": 38, 471 | "outputs": [] 472 | }, 473 | { 474 | "cell_type": "code", 475 | "source": [ 476 | "#Probabilidade de ocorrer A e B Multualmente\n", 477 | "univers.adicao_multual()" 478 | ], 479 | "metadata": { 480 | "colab": { 481 | "base_uri": "https://localhost:8080/" 482 | }, 483 | "id": "VsE_z-mtu095", 484 | "outputId": "3014cd48-d03e-42b2-880b-3a12382c41f5" 485 | }, 486 | "execution_count": 39, 487 | "outputs": [ 488 | { 489 | "output_type": "execute_result", 490 | "data": { 491 | "text/plain": [ 492 | "0.6" 493 | ] 494 | }, 495 | "metadata": {}, 496 | "execution_count": 39 497 | } 498 | ] 499 | }, 500 | { 501 | "cell_type": "markdown", 502 | "source": [ 503 | "### Exemplo 5" 504 | ], 505 | "metadata": { 506 | "id": "khllTAi8I_8z" 507 | } 508 | }, 509 | { 510 | "cell_type": "code", 511 | "source": [ 512 | "#E) probobalidade de ocorrer mas A e B são mutuamente excludentes\n", 513 | "univers.adicao_excludantes()" 514 | ], 515 | "metadata": { 516 | "colab": { 517 | "base_uri": "https://localhost:8080/" 518 | }, 519 | "id": "NYoeUCVbi2Ls", 520 | "outputId": "9f2ab0a4-7128-475a-efb5-8d0d3a779c89" 521 | }, 522 | "execution_count": 40, 523 | "outputs": [ 524 | { 525 | "output_type": "execute_result", 526 | "data": { 527 | "text/plain": [ 528 | "0.7" 529 | ] 530 | }, 531 | "metadata": {}, 532 | "execution_count": 40 533 | } 534 | ] 535 | }, 536 | { 537 | "cell_type": "markdown", 538 | "source": [ 539 | "### Exemplo 6" 540 | ], 541 | "metadata": { 542 | "id": "QAncuLPNJBM2" 543 | } 544 | }, 545 | { 546 | "cell_type": "code", 547 | "source": [ 548 | "#F) Probabilidade de eventos complementares\n", 549 | "univers.complement()" 550 | ], 551 | "metadata": { 552 | "colab": { 553 | "base_uri": "https://localhost:8080/" 554 | }, 555 | "id": "RPeFdsDOnE1k", 556 | "outputId": "9ef669e5-86d8-4614-db4f-cc15ca3e27cc" 557 | }, 558 | "execution_count": 41, 559 | "outputs": [ 560 | { 561 | "output_type": "execute_result", 562 | "data": { 563 | "text/plain": [ 564 | "0.5" 565 | ] 566 | }, 567 | "metadata": {}, 568 | "execution_count": 41 569 | } 570 | ] 571 | }, 572 | { 573 | "cell_type": "markdown", 574 | "source": [ 575 | "### Exemplo 7" 576 | ], 577 | "metadata": { 578 | "id": "_MGzbx9cJCYP" 579 | } 580 | }, 581 | { 582 | "cell_type": "code", 583 | "source": [ 584 | "#G) Probabilidade de eventos independentes\n", 585 | "univers.indepedent()" 586 | ], 587 | "metadata": { 588 | "colab": { 589 | "base_uri": "https://localhost:8080/" 590 | }, 591 | "id": "9qcrKRqhnUgG", 592 | "outputId": "d3e668e9-0ff9-4572-b0b5-30c81f4780de" 593 | }, 594 | "execution_count": 42, 595 | "outputs": [ 596 | { 597 | "output_type": "execute_result", 598 | "data": { 599 | "text/plain": [ 600 | "0.1" 601 | ] 602 | }, 603 | "metadata": {}, 604 | "execution_count": 42 605 | } 606 | ] 607 | }, 608 | { 609 | "cell_type": "markdown", 610 | "source": [ 611 | "## **Exercícios Probabilidade Condicional**" 612 | ], 613 | "metadata": { 614 | "id": "3xUhN4oJJNYM" 615 | } 616 | }, 617 | { 618 | "cell_type": "markdown", 619 | "source": [ 620 | "### Exemplo 1" 621 | ], 622 | "metadata": { 623 | "id": "xo-0TTs5KSGW" 624 | } 625 | }, 626 | { 627 | "cell_type": "code", 628 | "source": [ 629 | "# Probabilidade da condição A ocorrer B\n", 630 | "univers.conditional()" 631 | ], 632 | "metadata": { 633 | "colab": { 634 | "base_uri": "https://localhost:8080/" 635 | }, 636 | "id": "BoQw_EsOJzlU", 637 | "outputId": "395a9122-e3c4-4611-8812-fb5c90e575bb" 638 | }, 639 | "execution_count": 43, 640 | "outputs": [ 641 | { 642 | "output_type": "execute_result", 643 | "data": { 644 | "text/plain": [ 645 | "0.5" 646 | ] 647 | }, 648 | "metadata": {}, 649 | "execution_count": 43 650 | } 651 | ] 652 | }, 653 | { 654 | "cell_type": "markdown", 655 | "source": [ 656 | "### Exemplo 2" 657 | ], 658 | "metadata": { 659 | "id": "YHlRGUUEKU7g" 660 | } 661 | }, 662 | { 663 | "cell_type": "code", 664 | "source": [ 665 | "# Probabilidade da condição de A B e C ocorrer\n", 666 | "C = nao_mult_cinco(S)\n", 667 | "univers2 = universo(A, S, B, C)\n", 668 | "univers2.conditional_mult()" 669 | ], 670 | "metadata": { 671 | "colab": { 672 | "base_uri": "https://localhost:8080/" 673 | }, 674 | "id": "BWprxn_xKVpQ", 675 | "outputId": "949fcab3-15a1-4dd7-94f9-14d000c8ed8b" 676 | }, 677 | "execution_count": 44, 678 | "outputs": [ 679 | { 680 | "output_type": "execute_result", 681 | "data": { 682 | "text/plain": [ 683 | "0.03125" 684 | ] 685 | }, 686 | "metadata": {}, 687 | "execution_count": 44 688 | } 689 | ] 690 | } 691 | ] 692 | } -------------------------------------------------------------------------------- /cap7_teste_hipotese.ipynb: -------------------------------------------------------------------------------- 1 | { 2 | "nbformat": 4, 3 | "nbformat_minor": 0, 4 | "metadata": { 5 | "colab": { 6 | "provenance": [] 7 | }, 8 | "kernelspec": { 9 | "name": "python3", 10 | "display_name": "Python 3" 11 | }, 12 | "accelerator": "GPU" 13 | }, 14 | "cells": [ 15 | { 16 | "cell_type": "markdown", 17 | "metadata": { 18 | "id": "kLI0gJtzDGuQ" 19 | }, 20 | "source": [ 21 | "**by: [João Ataíde](https://www.joaoataide.com)**\n", 22 | "# **CAP 07: Teste Hipótese**\n", 23 | "\n", 24 | "---\n", 25 | "\n" 26 | ] 27 | }, 28 | { 29 | "cell_type": "code", 30 | "source": [ 31 | "#importar bibliotecas\n", 32 | "import pandas as pd\n", 33 | "import numpy as np\n", 34 | "import math\n", 35 | "import random\n", 36 | "import matplotlib.pyplot as plt\n", 37 | "import seaborn as sns\n", 38 | "import statistics as st\n", 39 | "import scipy.stats as stats\n", 40 | "from scipy.stats import norm\n", 41 | "\n", 42 | "from sklearn.model_selection import train_test_split\n", 43 | "from scipy.stats import chi2_contingency, spearmanr\n", 44 | "\n", 45 | "import matplotlib\n", 46 | "matplotlib.style.use('dark_background')\n", 47 | "matplotlib.rcParams['figure.figsize'] = (6.0, 6.0)\n", 48 | "matplotlib.rcParams['image.cmap'] = 'gray'\n", 49 | "%matplotlib inline" 50 | ], 51 | "metadata": { 52 | "id": "dTSRYPl7ORJW" 53 | }, 54 | "execution_count": 35, 55 | "outputs": [] 56 | }, 57 | { 58 | "cell_type": "markdown", 59 | "source": [ 60 | "Um problema das inferências estatísticas é o teste de hipótese, esse termo `hipótese estatítica` é uma \"Suposição\" sobre determinado parâmetros da população, como média, desvio, correlação e outros.\n", 61 | "\n", 62 | "O `o teste de hipótese` é exatamente o procedimento para identificar a veracidade ou falsidade de determinada hipótese.\n", 63 | "\n", 64 | "\n", 65 | "Uma variável $X$ associada a um determinado parâmetros $Θ$.\n", 66 | "\n", 67 | "- Determinar a hipótese a ser analisada sobre o parâmetros (`Hipótese Nula`)\n", 68 | "$H_0: θ = θ_0\n", 69 | "$\n", 70 | "\n", 71 | "- E em seguida a hipótese alternativa $H_1$" 72 | ], 73 | "metadata": { 74 | "id": "QqttOxDHRids" 75 | } 76 | }, 77 | { 78 | "cell_type": "markdown", 79 | "source": [ 80 | "## **Teste unilateral à esquerda ou unicaldal**\n", 81 | "\n", 82 | "$α$ é o nível de significância, basicamente é a probabilidade de rejeitar uma hipótese nula quando ela for verdadeira, podendo ter dois tipos de erro `RN` Região de Não Regeião e `RC` Região Crítica.\n", 83 | "\n", 84 | "Hipotese alternativa seria,$H_1: θ \\neq θ_0$.\n", 85 | "\n", 86 | "Veja como seria o gráfico desse teste.\n", 87 | "\n" 88 | ], 89 | "metadata": { 90 | "id": "vyaGguhnWIJg" 91 | } 92 | }, 93 | { 94 | "cell_type": "code", 95 | "source": [ 96 | "# Parâmetros\n", 97 | "mu = 0\n", 98 | "variance = 1\n", 99 | "sigma = np.sqrt(variance)\n", 100 | "x = np.linspace(mu - 3.5*sigma, mu + 3.5*sigma, 100)\n", 101 | "alpha = 0.05" 102 | ], 103 | "metadata": { 104 | "id": "rfEJBeOLs1IK" 105 | }, 106 | "execution_count": 36, 107 | "outputs": [] 108 | }, 109 | { 110 | "cell_type": "code", 111 | "source": [ 112 | "# Cria a distribuição normal\n", 113 | "plt.plot(x, norm.pdf(x, mu, sigma), color = 'white')\n", 114 | "\n", 115 | "# Adiciona a região de rejeição\n", 116 | "low = norm.ppf(alpha)\n", 117 | "high = norm.ppf(1 - alpha)\n", 118 | "\n", 119 | "# Adiciona a região de não rejeição\n", 120 | "plt.fill_between(x, norm.pdf(x, mu, sigma), where=(x<=low) | (x>=high), color='red', alpha=0.5)\n", 121 | "\n", 122 | "# Adiciona a anotação\n", 123 | "plt.text(low, norm.pdf(low, mu, sigma), f'alpha/2', ha='right')\n", 124 | "plt.text(high, norm.pdf(high, mu, sigma), f'alpha/2', ha='left')\n", 125 | "\n", 126 | "# Adiciona as legendas\n", 127 | "plt.legend([\"Distribuição\",'RC', 'RN'])\n", 128 | "\n", 129 | "# Mostra o gráfico\n", 130 | "plt.show()" 131 | ], 132 | "metadata": { 133 | "colab": { 134 | "base_uri": "https://localhost:8080/", 135 | "height": 522 136 | }, 137 | "id": "012WNqrJzJh7", 138 | "outputId": "6b6f485a-bd24-4746-89b5-24615e990a6c" 139 | }, 140 | "execution_count": 99, 141 | "outputs": [ 142 | { 143 | "output_type": "display_data", 144 | "data": { 145 | "text/plain": [ 146 | "
" 147 | ], 148 | "image/png": "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\n" 149 | }, 150 | "metadata": {} 151 | } 152 | ] 153 | }, 154 | { 155 | "cell_type": "markdown", 156 | "source": [ 157 | "## **Teste unilateral ou unicaldal**\n", 158 | "\n", 159 | "Usado para ver se o parâmetro e significativamente superior ou inferiro ao valor.\n", 160 | "\n", 161 | "Hipotese alternativa a esquerda seria,\n", 162 | "\n", 163 | "$H_1: θ < θ_0\n", 164 | "$\n", 165 | "\n" 166 | ], 167 | "metadata": { 168 | "id": "_h0Ab9-kTayN" 169 | } 170 | }, 171 | { 172 | "cell_type": "code", 173 | "source": [ 174 | "# Cria a distribuição normal\n", 175 | "plt.plot(x, norm.pdf(x, mu, sigma), color = 'white')\n", 176 | "\n", 177 | "# Adiciona a região de rejeição\n", 178 | "low = norm.ppf(alpha)\n", 179 | "plt.fill_between(x, norm.pdf(x, mu, sigma), where=(x<=low), color='red', alpha=0.5)\n", 180 | "\n", 181 | "# Adiciona a anotação\n", 182 | "plt.text(low, norm.pdf(low, mu, sigma), f'alpha', ha='right')\n", 183 | "\n", 184 | "# Adiciona as legendas\n", 185 | "plt.legend([\"Distribuição\",'RC', 'RN'])\n", 186 | "\n", 187 | "# Mostra o gráfico\n", 188 | "plt.show()" 189 | ], 190 | "metadata": { 191 | "colab": { 192 | "base_uri": "https://localhost:8080/", 193 | "height": 522 194 | }, 195 | "id": "mu7NBhZ0WQ_5", 196 | "outputId": "2f1053a8-ac3c-49a9-b25c-c9728d99e283" 197 | }, 198 | "execution_count": 95, 199 | "outputs": [ 200 | { 201 | "output_type": "display_data", 202 | "data": { 203 | "text/plain": [ 204 | "
" 205 | ], 206 | "image/png": "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\n" 207 | }, 208 | "metadata": {} 209 | } 210 | ] 211 | }, 212 | { 213 | "cell_type": "markdown", 214 | "source": [ 215 | "Hipotese alternativa a direita seria,\n", 216 | "\n", 217 | "$H_1: θ > θ_0\n", 218 | "$" 219 | ], 220 | "metadata": { 221 | "id": "vJl2JUdaY1DL" 222 | } 223 | }, 224 | { 225 | "cell_type": "code", 226 | "source": [ 227 | "# Cria a distribuição normal\n", 228 | "plt.plot(x, norm.pdf(x, mu, sigma), color = 'white')\n", 229 | "\n", 230 | "# Adiciona a região de rejeição\n", 231 | "high = norm.ppf(1 - alpha)\n", 232 | "plt.fill_between(x, norm.pdf(x, mu, sigma), where=(x>high), color='red', alpha=0.5)\n", 233 | "\n", 234 | "# Adiciona a anotação\n", 235 | "plt.text(high, norm.pdf(high, mu, sigma), f'alpha', ha='left')\n", 236 | "\n", 237 | "# Adiciona as legendas\n", 238 | "plt.legend([\"Distribuição\",'RC', 'RN'])\n", 239 | "\n", 240 | "# Mostra o gráfico\n", 241 | "plt.show()" 242 | ], 243 | "metadata": { 244 | "colab": { 245 | "base_uri": "https://localhost:8080/", 246 | "height": 522 247 | }, 248 | "id": "j8QlygnsWRdN", 249 | "outputId": "0499eca8-a12f-483b-983c-ce748ea5aaa7" 250 | }, 251 | "execution_count": 100, 252 | "outputs": [ 253 | { 254 | "output_type": "display_data", 255 | "data": { 256 | "text/plain": [ 257 | "
" 258 | ], 259 | "image/png": "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\n" 260 | }, 261 | "metadata": {} 262 | } 263 | ] 264 | }, 265 | { 266 | "cell_type": "markdown", 267 | "source": [ 268 | "## **Tipos de Erros**" 269 | ], 270 | "metadata": { 271 | "id": "YJPXaTX7W9cU" 272 | } 273 | }, 274 | { 275 | "cell_type": "markdown", 276 | "source": [ 277 | "Definida as duas hipóteses usanod uma amostra probabilistica comprovamos ou não as hipóteses, gerando então dois tipos de erros:\n", 278 | "\n", 279 | "- `Erro Tipo I` Rejeita a Hipóitese Nula quando for verdadeira\n", 280 | "\n", 281 | "$P(E_I) = P(\\text{rejeitar } H_0 \\text{ | } H_0 \\text{ é verdadeira}) = \\alpha$\n", 282 | "\n", 283 | "- `Erro Tipo II` Não Rejeita a hipótese nula quando for falsa.\n", 284 | "\n", 285 | "$P(E_{II}) = P(\\text{rejeitar } H_0 \\text{ | } H_0 \\text{ é falsa}) = \\beta$\n" 286 | ], 287 | "metadata": { 288 | "id": "Kwi3npBYXBAj" 289 | } 290 | } 291 | ] 292 | } --------------------------------------------------------------------------------