3.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Лабораторная работа №3. Однофакторный дисперсионный анализ\n",
    "\n",
    "> Вариант № ??\n",
    "\n",
    "**Распределения**:\n",
    "\n",
    "$X_1$ ~ ?? (объём выборки $n_1$ &mdash; ?)\n",
    "\n",
    "$X_2$ ~ ?? (объём выборки $n_2$ &mdash; ?)\n",
    "\n",
    "$X_3$ ~ ?? (объём выборки $n_3$ &mdash; ?)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "%matplotlib inline\n",
    "import numpy as np\n",
    "from scipy import stats\n",
    "import matplotlib.pyplot as plt\n",
    "from statsmodels.distributions.empirical_distribution import ECDF\n",
    "\n",
    "# Немного магии для того, чтобы рисунки стали больше\n",
    "import pylab\n",
    "pylab.rcParams['figure.figsize'] = (24.0, 16.0)\n",
    "plt.rcParams.update({'font.size': 22})\n",
    "\n",
    "# Вспомогательные классы и функции, определённые тут же (см. репозиторий)\n",
    "# https://github.com/BobNobrain/matstat-labs/tree/master/s\n",
    "from s import Sample, PooledSample, BartlettHyp, OneWayAnovaHyp, LinearContrastHyp\n",
    "from s.utils import table, printf"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 1. Исходные данные"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Характеристики наблюдаемых случайных величин:\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<table width=\"100%\">\n",
       "    <thead><tr><td>СВ</td><td>Распределение</td><td>Параметры</td><td>$m_i$</td><td>$\\sigma_i^2$</td><td>$n_i$</td></tr></thead>\n",
       "    <tbody><tr><td>X1</td><td>N</td><td>$m=0.0, \\sigma=1.0$</td><td>0.0</td><td>1.0</td><td>100</td></tr><tr><td>X2</td><td>N</td><td>$m=0.1, \\sigma=1.1$</td><td>0.01</td><td>1.2100000000000002</td><td>200</td></tr><tr><td>X3</td><td>N</td><td>$m=-0.3, \\sigma=0.9$</td><td>-0.02</td><td>0.81</td><td>150</td></tr></tbody>\n",
       "    </table>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<p>Количество случайных величин $k=3$</p>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "X1 = stats.norm(0, 1)\n",
    "X2 = stats.norm(0.01, 1.1)\n",
    "X3 = stats.norm(-0.02, 0.9)\n",
    "\n",
    "n1 = 100\n",
    "n2 = 200\n",
    "n3 = 150\n",
    "\n",
    "print('Характеристики наблюдаемых случайных величин:')\n",
    "table(\n",
    "    ['СВ', 'Распределение', 'Параметры', '$m_i$', '$\\\\sigma_i^2$', '$n_i$'],\n",
    "    [\n",
    "        ['X1', 'N', '$m=0.0, \\\\sigma=1.0$', X1.mean(), X1.var(), n1],\n",
    "        ['X2', 'N', '$m=0.1, \\\\sigma=1.1$', X2.mean(), X2.var(), n2],\n",
    "        ['X3', 'N', '$m=-0.3, \\\\sigma=0.9$', X3.mean(), X3.var(), n3]\n",
    "    ]\n",
    ")\n",
    "printf('Количество случайных величин $k={}$', 3)\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Выборочные характеристики:\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<table width=\"100%\">\n",
       "    <thead><tr><td>СВ</td><td>Среднее $\\overline{x_i}$</td><td>$s^2_i$</td><td>$s_i$</td></tr></thead>\n",
       "    <tbody><tr><td>x1</td><td>-0.0146</td><td>0.8012</td><td>0.8951</td></tr><tr><td>x2</td><td>0.0003</td><td>1.0764</td><td>1.0375</td></tr><tr><td>x3</td><td>-0.1252</td><td>0.77</td><td>0.8775</td></tr><tr><td>Pooled</td><td>-0.0448</td><td>0.9125</td><td>0.9553</td></tr></tbody>\n",
       "    </table>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "x1 = Sample.from_distribution(\"x1\", X1, count=n1)\n",
    "x2 = Sample.from_distribution(\"x2\", X2, count=n2)\n",
    "x3 = Sample.from_distribution(\"x3\", X3, count=n3)\n",
    "\n",
    "x_pooled = PooledSample(\"Pooled\", x1, x2, x3)\n",
    "\n",
    "print('Выборочные характеристики:')\n",
    "table(\n",
    "    ['СВ', 'Среднее $\\\\overline{x_i}$', '$s^2_i$', '$s_i$'],\n",
    "    [\n",
    "        [\n",
    "            s._name,\n",
    "            round(s.mean(), 4),\n",
    "            round(s.s() ** 2, 4),\n",
    "            round(s.s(), 4)\n",
    "        ] for s in [x1, x2, x3, x_pooled]\n",
    "    ]\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 2. Визуальное представление выборок\n",
    "\n",
    "Диаграммы *Box-and-Whisker*:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7facd9f93898>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots()\n",
    "\n",
    "ax.boxplot([x1.data(), x2.data(), x3.data()])\n",
    "ax.set_title('Выборки')\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 3. Проверка условия применимости дисперсионного анализа\n",
    "\n",
    "Статистическая гипотеза $H_0: \\sigma_1^2 = \\sigma_2^2 = \\sigma_3^2$"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Критерий Бартлетта\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<table width=\"100%\">\n",
       "    <thead><tr><td>Выборочное значение статистики критерия</td><td>p-value</td><td>Статистическое решение при $\\alpha=0.05$</td><td>Ошибка статистического решения</td></tr></thead>\n",
       "    <tbody><tr><td>5.6829</td><td>0.0583</td><td>H0</td><td>TODO</td></tr></tbody>\n",
       "    </table>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "print('Критерий Бартлетта')\n",
    "\n",
    "alpha = 0.05\n",
    "\n",
    "H0 = BartlettHyp(x_pooled)\n",
    "criterion_value, interval, p_value, result = H0.full_test(x_pooled, alpha)\n",
    "\n",
    "table(\n",
    "    [\n",
    "        'Выборочное значение статистики критерия',\n",
    "        'p-value',\n",
    "        'Статистическое решение при $\\\\alpha={}$'.format(alpha),\n",
    "        'Ошибка статистического решения'\n",
    "    ],\n",
    "    [[\n",
    "        round(criterion_value, 4),\n",
    "        round(p_value, 4),\n",
    "        'H0' if result else 'H1',\n",
    "        'TODO'\n",
    "    ]]\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 4. Однофакторный дисперсионный анализ"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Таблица дисперсионного анализа\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<table width=\"100%\">\n",
       "    <thead><tr><td>Источник вариации</td><td>Показатель вариации</td><td>Число степеней свободы</td><td>Несмещённая оценка</td></tr></thead>\n",
       "    <tbody><tr><td>Группировочный признак</td><td></td><td></td><td></td></tr><tr><td>Остаточные признаки</td><td></td><td></td><td></td></tr><tr><td>Все признаки</td><td></td><td></td><td></td></tr></tbody>\n",
       "    </table>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<p>Эмпирический коэффициент детерминации $\\eta^2 = 0.0054$</p>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<p>Эмпирическое корреляционное отношение $\\eta = 0.0738$</p>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "print('Таблица дисперсионного анализа')\n",
    "\n",
    "# http://datalearning.ru/index.php/textbook?cid=1&mid=5&topic=2, таблица 6.4\n",
    "table(\n",
    "    ['Источник вариации', 'Показатель вариации', 'Число степеней свободы', 'Несмещённая оценка'],\n",
    "    [\n",
    "        ['Группировочный признак', '', '', ''],\n",
    "        ['Остаточные признаки', '', '', ''],\n",
    "        ['Все признаки', '', '', '']\n",
    "    ]\n",
    ")\n",
    "\n",
    "eta2 = x_pooled.eta_squared()\n",
    "printf('Эмпирический коэффициент детерминации $\\\\eta^2 = {}$', round(eta2, 4))\n",
    "printf('Эмпирическое корреляционное отношение $\\\\eta = {}$', round(np.sqrt(eta2), 4))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Статистическая гипотеза $H_0: m_1 = m_2 = m_3$"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table width=\"100%\">\n",
       "    <thead><tr><td>Выборочное значение статистики критерия</td><td>p-value</td><td>Статистическое решение при $\\alpha=0.05$</td><td>Ошибка статистического решения</td></tr></thead>\n",
       "    <tbody><tr><td>4.2893</td><td>0.0143</td><td>H1</td><td>TODO</td></tr></tbody>\n",
       "    </table>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "alpha = 0.05\n",
    "\n",
    "anova = OneWayAnovaHyp(x_pooled)\n",
    "criterion_value, (crit_left, crit_right), p_value, result = anova.full_test(x_pooled, alpha)\n",
    "\n",
    "table(\n",
    "    [\n",
    "        'Выборочное значение статистики критерия',\n",
    "        'p-value',\n",
    "        'Статистическое решение при $\\\\alpha={}$'.format(alpha),\n",
    "        'Ошибка статистического решения'\n",
    "    ],\n",
    "    [[\n",
    "        round(criterion_value, 4),\n",
    "        round(p_value, 4),\n",
    "        'H0' if result else 'H1',\n",
    "        'TODO'\n",
    "    ]]\n",
    ")\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 5. Метод линейных контрастов"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7facda0179b0>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "-0.0524255939271 0.0225566655633\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<table width=\"100%\">\n",
       "    <thead><tr><td>Гипотеза</td><td>Выборочное значение статистики критерия</td><td>p-value</td><td>Статистическое решение при $\\alpha=0.05$</td><td>Ошибка статистического решения</td></tr></thead>\n",
       "    <tbody><tr><td>$m_1=m_2$</td><td>-0.0149</td><td>0.0</td><td>$=$</td><td>TODO</td></tr><tr><td>$m_2=m_3$</td><td></td><td></td><td></td><td></td></tr><tr><td>$m_1=m_3$</td><td></td><td></td><td></td><td></td></tr></tbody>\n",
       "    </table>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "alpha = 0.05\n",
    "\n",
    "def m_interval(sample):\n",
    "    n = sample.n()\n",
    "    delta = stats.t(n - 1).ppf(1 - alpha / 2) * sample.s() / np.sqrt(n)\n",
    "    mean = sample.mean()\n",
    "    return mean - delta, mean + delta\n",
    "    \n",
    "fig, ax = plt.subplots()\n",
    "ax.set_title('Доверительные интервалы для $m_{1..k}$')\n",
    "\n",
    "samples = [x1, x2, x3]\n",
    "for i in range(len(samples)):\n",
    "    l, r = m_interval(samples[i])\n",
    "    domain = [l, r]\n",
    "    values = [i + 1, i + 1]\n",
    "    ax.plot(\n",
    "        domain,\n",
    "        values,\n",
    "        label='$m_{} \\\\in [{}; {}]$'.format(i + 1, round(l, 3), round(r, 3)),\n",
    "        linewidth=4\n",
    "    )\n",
    "    ax.fill_between(domain, 0, values, alpha=.2)\n",
    "\n",
    "plt.legend()\n",
    "plt.show()\n",
    "\n",
    "# TODO: вынести в функцию и сделать для всех комбинаций\n",
    "H0 = LinearContrastHyp(x_pooled)\n",
    "c, (c_1, c_2), p_value, result = H0.full_test([1, -1, 0], alpha=alpha)\n",
    "print(c_1, c_2)\n",
    "\n",
    "table(\n",
    "    [\n",
    "        'Гипотеза',\n",
    "        'Выборочное значение статистики критерия',\n",
    "        'p-value',\n",
    "        'Статистическое решение при $\\\\alpha={}$'.format(alpha),\n",
    "        'Ошибка статистического решения'\n",
    "    ],\n",
    "    [\n",
    "        ['$m_1=m_2$', round(c, 4), round(p_value, 4), '$=$' if result else '$\\\\ne$', 'TODO'],\n",
    "        ['$m_2=m_3$', '', '', '', ''],\n",
    "        ['$m_1=m_3$', '', '', '', '']\n",
    "    ]\n",
    ")"
   ]
  }
 ],
 "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": "3.5.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}