lecture12.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "1d89b963-06e9-4d46-9654-e259e369d957",
   "metadata": {},
   "source": [
    "# 第１２回　多次元データの解析　（１）相関分析・回帰分析"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "40ac8f60-b217-4577-80e7-2154d1b701a5",
   "metadata": {},
   "source": [
    "今回は、次のような内容を取扱います。  \n",
    "- 相関について\n",
    "- 相関分析、相関係数\n",
    "- 回帰分析（単回帰分析、重回帰分析）\n",
    "- （補）分散分析"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2cb7b8bb-ebf1-4079-849f-a5047d08ada7",
   "metadata": {},
   "source": [
    "また、今回はとくに次のツールを使います。\n",
    "- 散布図マトリクスを描くためにmlxtendを（参考として）。\n",
    "- 散布図マトリクスを描くためにseabornを。\n",
    "- 線形回帰を行うためにstatsmodelsを。\n",
    "- 残差診断のためにlmdiagを（参考として）。"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "91c042d0-eb70-4b26-98d5-a1b5fb4581b0",
   "metadata": {},
   "source": [
    "## 相関について"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "04a795f1-fd93-475a-90d3-b3de4ecd5ed9",
   "metadata": {
    "tags": []
   },
   "source": [
    "### 散布図を描く"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d9ce7d75-8494-4d0c-964e-44d8567fa000",
   "metadata": {},
   "source": [
    "散布図は相関(correlation)を調べるための最初のステップとして重要です。  \n",
    "散布図を描いてみましょう。  \n",
    "ここでは、次のことをおこないます。\n",
    "- Numpy とPandasのインポート\n",
    "- データフレームとしてCSVファイルを読み込む\n",
    "- matplotlibのインライン宣言とインポート\n",
    "- mlextendを利用した散布図の描画（参考）\n",
    "- seabornによる散布図の描画"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e216fae0-9af0-4174-aae2-ccbea47cfc03",
   "metadata": {},
   "source": [
    "まず、NumPyとPandasのインポートをおこないます。今回主に使うのはPandasのほうです。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "940d5f10-650b-47ee-9d19-c80d0f9da63a",
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "43391eb0-7d75-4f51-a937-1a2ee20af427",
   "metadata": {},
   "source": [
    "日本統計学会編「改訂版　統計学基礎」（東京図書）の第１章で使われる家賃データをpandas.read_csv()関数を用いてCSVファイルから読み込みます。  \n",
    "東京図書のWebサイトからダウンロードしたファイルの先頭行はそれぞれの部屋の属性を表します。すなわち、駅からの近さ、家賃、間取り、広さ、部屋の向き、そして築年数です。読み込むファイルではこれらは英語に直してあります。  \n",
    "また、保存するファイルのエンコーディング形式はUTF-8 (別名で保存するときに、UTF-8 CSV形式を選びます）としました。もともとはShift-JIS形式で作られていました。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "999bde86-af1c-4123-8567-d32cdc5ca73c",
   "metadata": {},
   "outputs": [],
   "source": [
    "df_room = pd.read_csv('room-utf8.csv')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "401d38f7-63fd-48c2-8d64-98dc53bcab9f",
   "metadata": {},
   "source": [
    "読み込むと、「**データフレーム**」ができます。第４回で説明したように、データフレームでは各列に項目名でアクセスできます。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "22938dc7-43ed-4131-afad-93ba4d4bc507",
   "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>proximity</th>\n",
       "      <th>rent</th>\n",
       "      <th>plan</th>\n",
       "      <th>size</th>\n",
       "      <th>direction</th>\n",
       "      <th>age</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>B</td>\n",
       "      <td>68000</td>\n",
       "      <td>1K</td>\n",
       "      <td>19</td>\n",
       "      <td>W</td>\n",
       "      <td>12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>B</td>\n",
       "      <td>68000</td>\n",
       "      <td>1K</td>\n",
       "      <td>19</td>\n",
       "      <td>S</td>\n",
       "      <td>12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>B</td>\n",
       "      <td>69000</td>\n",
       "      <td>1K</td>\n",
       "      <td>19</td>\n",
       "      <td>NW</td>\n",
       "      <td>14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>B</td>\n",
       "      <td>70000</td>\n",
       "      <td>1K</td>\n",
       "      <td>19</td>\n",
       "      <td>S</td>\n",
       "      <td>14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>B</td>\n",
       "      <td>72000</td>\n",
       "      <td>1K</td>\n",
       "      <td>15</td>\n",
       "      <td>S</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>135</th>\n",
       "      <td>B</td>\n",
       "      <td>145000</td>\n",
       "      <td>1DK</td>\n",
       "      <td>40</td>\n",
       "      <td>S</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>136</th>\n",
       "      <td>B</td>\n",
       "      <td>145000</td>\n",
       "      <td>1SLDK</td>\n",
       "      <td>54</td>\n",
       "      <td>SE</td>\n",
       "      <td>28</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>137</th>\n",
       "      <td>A</td>\n",
       "      <td>148000</td>\n",
       "      <td>1LDK</td>\n",
       "      <td>42</td>\n",
       "      <td>SE</td>\n",
       "      <td>13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>138</th>\n",
       "      <td>A</td>\n",
       "      <td>148000</td>\n",
       "      <td>1LDK</td>\n",
       "      <td>42</td>\n",
       "      <td>S</td>\n",
       "      <td>13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>139</th>\n",
       "      <td>B</td>\n",
       "      <td>150000</td>\n",
       "      <td>1LDK</td>\n",
       "      <td>41</td>\n",
       "      <td>SE</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>140 rows × 6 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "    proximity    rent   plan  size direction  age\n",
       "0           B   68000     1K    19         W   12\n",
       "1           B   68000     1K    19         S   12\n",
       "2           B   69000     1K    19        NW   14\n",
       "3           B   70000     1K    19         S   14\n",
       "4           B   72000     1K    15         S    9\n",
       "..        ...     ...    ...   ...       ...  ...\n",
       "135         B  145000    1DK    40         S    8\n",
       "136         B  145000  1SLDK    54        SE   28\n",
       "137         A  148000   1LDK    42        SE   13\n",
       "138         A  148000   1LDK    42         S   13\n",
       "139         B  150000   1LDK    41        SE    5\n",
       "\n",
       "[140 rows x 6 columns]"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_room"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "241735f6-0cc8-4de3-bf0b-6c1eec6310c0",
   "metadata": {},
   "source": [
    "ここで、proximityの'A'は駅から7分まで、'B'は8分から15分までを表します。"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "dedfffec-11a0-4e0a-a1a0-d1bd8892bd50",
   "metadata": {},
   "source": [
    "どのような変数があるか、表示します。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "56bd9dfa-8ab3-41bc-846e-90b76cb42a47",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['proximity', 'rent', 'plan', 'size', 'direction', 'age'], dtype='object')"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_room.columns"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "59f9d604-87be-4a2c-b934-678b22dedf51",
   "metadata": {},
   "source": [
    "データセットの概要を表示します。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "id": "8706211c-2c0f-44ca-a823-d7e67b599e76",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 140 entries, 0 to 139\n",
      "Data columns (total 6 columns):\n",
      " #   Column     Non-Null Count  Dtype \n",
      "---  ------     --------------  ----- \n",
      " 0   proximity  140 non-null    object\n",
      " 1   rent       140 non-null    int64 \n",
      " 2   plan       140 non-null    object\n",
      " 3   size       140 non-null    int64 \n",
      " 4   direction  140 non-null    object\n",
      " 5   age        140 non-null    int64 \n",
      "dtypes: int64(3), object(3)\n",
      "memory usage: 6.7+ KB\n"
     ]
    }
   ],
   "source": [
    "df_room.info()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f0cc9616-9bd5-49dc-a4a1-611d95196ed4",
   "metadata": {},
   "source": [
    "先頭の10行を表示します。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "b0c5abe3-9790-4a2d-871d-eb3a4a7c8f24",
   "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>proximity</th>\n",
       "      <th>rent</th>\n",
       "      <th>plan</th>\n",
       "      <th>size</th>\n",
       "      <th>direction</th>\n",
       "      <th>age</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>B</td>\n",
       "      <td>68000</td>\n",
       "      <td>1K</td>\n",
       "      <td>19</td>\n",
       "      <td>W</td>\n",
       "      <td>12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>B</td>\n",
       "      <td>68000</td>\n",
       "      <td>1K</td>\n",
       "      <td>19</td>\n",
       "      <td>S</td>\n",
       "      <td>12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>B</td>\n",
       "      <td>69000</td>\n",
       "      <td>1K</td>\n",
       "      <td>19</td>\n",
       "      <td>NW</td>\n",
       "      <td>14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>B</td>\n",
       "      <td>70000</td>\n",
       "      <td>1K</td>\n",
       "      <td>19</td>\n",
       "      <td>S</td>\n",
       "      <td>14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>B</td>\n",
       "      <td>72000</td>\n",
       "      <td>1K</td>\n",
       "      <td>15</td>\n",
       "      <td>S</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>B</td>\n",
       "      <td>77000</td>\n",
       "      <td>1K</td>\n",
       "      <td>20</td>\n",
       "      <td>S</td>\n",
       "      <td>14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>B</td>\n",
       "      <td>77000</td>\n",
       "      <td>1K</td>\n",
       "      <td>20</td>\n",
       "      <td>W</td>\n",
       "      <td>14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>A</td>\n",
       "      <td>78000</td>\n",
       "      <td>1K</td>\n",
       "      <td>21</td>\n",
       "      <td>E</td>\n",
       "      <td>15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>B</td>\n",
       "      <td>79000</td>\n",
       "      <td>1K</td>\n",
       "      <td>22</td>\n",
       "      <td>S</td>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>B</td>\n",
       "      <td>79000</td>\n",
       "      <td>1K</td>\n",
       "      <td>22</td>\n",
       "      <td>S</td>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  proximity   rent plan  size direction  age\n",
       "0         B  68000   1K    19         W   12\n",
       "1         B  68000   1K    19         S   12\n",
       "2         B  69000   1K    19        NW   14\n",
       "3         B  70000   1K    19         S   14\n",
       "4         B  72000   1K    15         S    9\n",
       "5         B  77000   1K    20         S   14\n",
       "6         B  77000   1K    20         W   14\n",
       "7         A  78000   1K    21         E   15\n",
       "8         B  79000   1K    22         S   10\n",
       "9         B  79000   1K    22         S   10"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_room.head(10)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cea2d52e-245d-415a-ac62-6b715e1d084d",
   "metadata": {},
   "source": [
    "11行目から20行目までを表示しましょう。11行目のインデックス（添字）は10, 20行目は19になります。終わりよりも１多い数を終点に指定します。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "7cc4c402-e3a0-4330-bcc5-55ab534e536e",
   "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>proximity</th>\n",
       "      <th>rent</th>\n",
       "      <th>plan</th>\n",
       "      <th>size</th>\n",
       "      <th>direction</th>\n",
       "      <th>age</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>B</td>\n",
       "      <td>79000</td>\n",
       "      <td>1K</td>\n",
       "      <td>20</td>\n",
       "      <td>SE</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>A</td>\n",
       "      <td>80000</td>\n",
       "      <td>1K</td>\n",
       "      <td>20</td>\n",
       "      <td>NW</td>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>A</td>\n",
       "      <td>80000</td>\n",
       "      <td>1K</td>\n",
       "      <td>21</td>\n",
       "      <td>E</td>\n",
       "      <td>15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>B</td>\n",
       "      <td>80000</td>\n",
       "      <td>1K</td>\n",
       "      <td>22</td>\n",
       "      <td>E</td>\n",
       "      <td>14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>B</td>\n",
       "      <td>81000</td>\n",
       "      <td>1K</td>\n",
       "      <td>22</td>\n",
       "      <td>S</td>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>A</td>\n",
       "      <td>82000</td>\n",
       "      <td>1K</td>\n",
       "      <td>22</td>\n",
       "      <td>S</td>\n",
       "      <td>11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>A</td>\n",
       "      <td>82000</td>\n",
       "      <td>1K</td>\n",
       "      <td>21</td>\n",
       "      <td>E</td>\n",
       "      <td>15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>B</td>\n",
       "      <td>82000</td>\n",
       "      <td>1K</td>\n",
       "      <td>20</td>\n",
       "      <td>SE</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>A</td>\n",
       "      <td>83000</td>\n",
       "      <td>1K</td>\n",
       "      <td>22</td>\n",
       "      <td>S</td>\n",
       "      <td>13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>A</td>\n",
       "      <td>83000</td>\n",
       "      <td>1K</td>\n",
       "      <td>21</td>\n",
       "      <td>W</td>\n",
       "      <td>11</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   proximity   rent plan  size direction  age\n",
       "10         B  79000   1K    20        SE    9\n",
       "11         A  80000   1K    20        NW   10\n",
       "12         A  80000   1K    21         E   15\n",
       "13         B  80000   1K    22         E   14\n",
       "14         B  81000   1K    22         S   10\n",
       "15         A  82000   1K    22         S   11\n",
       "16         A  82000   1K    21         E   15\n",
       "17         B  82000   1K    20        SE    9\n",
       "18         A  83000   1K    22         S   13\n",
       "19         A  83000   1K    21         W   11"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_room[10:20]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ee3c0220-ba91-4de1-aabe-9d7eb3b4e969",
   "metadata": {},
   "source": [
    "散布図を書くために、matplotlibの準備をします。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "94c2a1fc-db1e-4a46-a81a-a6d37e54a7dd",
   "metadata": {},
   "outputs": [],
   "source": [
    "%matplotlib inline\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f85214bc-77da-4b9e-b18a-04919e8c9b56",
   "metadata": {},
   "source": [
    "散布図行列をプロットするために、まず、**mlxtend**を使ってみます。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "64c4cb96-4e32-43f1-b0da-f05b9877da30",
   "metadata": {},
   "outputs": [
    {
     "ename": "ModuleNotFoundError",
     "evalue": "No module named 'mlxtend'",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mModuleNotFoundError\u001b[0m                       Traceback (most recent call last)",
      "\u001b[1;32m~\\AppData\\Local\\Temp/ipykernel_10120/3487681511.py\u001b[0m in \u001b[0;36m<module>\u001b[1;34m\u001b[0m\n\u001b[1;32m----> 1\u001b[1;33m \u001b[1;32mfrom\u001b[0m \u001b[0mmlxtend\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mplotting\u001b[0m \u001b[1;32mimport\u001b[0m \u001b[0mscatterplotmatrix\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m",
      "\u001b[1;31mModuleNotFoundError\u001b[0m: No module named 'mlxtend'"
     ]
    }
   ],
   "source": [
    "from mlxtend.plotting import scatterplotmatrix"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "dc146acf-fffb-4703-8fd6-ba85a4e1eb57",
   "metadata": {},
   "source": [
    "pip installでインストールします。**注意：今回は散布図マトリクスを描くためにseabornを使いますのでmlxtendをインストールしなくて結構です。**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "68005c83-853e-4802-ae57-62cb8cb6839a",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Requirement already satisfied: mlxtend in e:\\pythonusb\\pythonusb\\python\\python-3.9.9\\lib\\site-packages (0.19.0)\n",
      "Requirement already satisfied: pandas>=0.24.2 in e:\\pythonusb\\pythonusb\\python\\python-3.9.9\\lib\\site-packages (from mlxtend) (1.4.2)\n",
      "Requirement already satisfied: joblib>=0.13.2 in e:\\pythonusb\\pythonusb\\python\\python-3.9.9\\lib\\site-packages (from mlxtend) (1.1.0)\n",
      "Requirement already satisfied: matplotlib>=3.0.0 in e:\\pythonusb\\pythonusb\\python\\python-3.9.9\\lib\\site-packages (from mlxtend) (3.5.0)\n",
      "Requirement already satisfied: scikit-learn>=0.20.3 in e:\\pythonusb\\pythonusb\\python\\python-3.9.9\\lib\\site-packages (from mlxtend) (1.1.1)\n",
      "Requirement already satisfied: setuptools in e:\\pythonusb\\pythonusb\\python\\python-3.9.9\\lib\\site-packages (from mlxtend) (59.2.0)\n",
      "Requirement already satisfied: scipy>=1.2.1 in e:\\pythonusb\\pythonusb\\python\\python-3.9.9\\lib\\site-packages (from mlxtend) (1.7.3)\n",
      "Requirement already satisfied: numpy>=1.16.2 in e:\\pythonusb\\pythonusb\\python\\python-3.9.9\\lib\\site-packages (from mlxtend) (1.21.4)\n",
      "Requirement already satisfied: pyparsing>=2.2.1 in e:\\pythonusb\\pythonusb\\python\\python-3.9.9\\lib\\site-packages (from matplotlib>=3.0.0->mlxtend) (3.0.6)\n",
      "Requirement already satisfied: pillow>=6.2.0 in e:\\pythonusb\\pythonusb\\python\\python-3.9.9\\lib\\site-packages (from matplotlib>=3.0.0->mlxtend) (8.4.0)\n",
      "Requirement already satisfied: cycler>=0.10 in e:\\pythonusb\\pythonusb\\python\\python-3.9.9\\lib\\site-packages (from matplotlib>=3.0.0->mlxtend) (0.11.0)\n",
      "Requirement already satisfied: python-dateutil>=2.7 in e:\\pythonusb\\pythonusb\\python\\python-3.9.9\\lib\\site-packages (from matplotlib>=3.0.0->mlxtend) (2.8.2)\n",
      "Requirement already satisfied: setuptools-scm>=4 in e:\\pythonusb\\pythonusb\\python\\python-3.9.9\\lib\\site-packages (from matplotlib>=3.0.0->mlxtend) (6.3.2)\n",
      "Requirement already satisfied: fonttools>=4.22.0 in e:\\pythonusb\\pythonusb\\python\\python-3.9.9\\lib\\site-packages (from matplotlib>=3.0.0->mlxtend) (4.28.2)\n",
      "Requirement already satisfied: packaging>=20.0 in e:\\pythonusb\\pythonusb\\python\\python-3.9.9\\lib\\site-packages (from matplotlib>=3.0.0->mlxtend) (21.3)\n",
      "Requirement already satisfied: kiwisolver>=1.0.1 in e:\\pythonusb\\pythonusb\\python\\python-3.9.9\\lib\\site-packages (from matplotlib>=3.0.0->mlxtend) (1.3.2)\n",
      "Requirement already satisfied: pytz>=2020.1 in e:\\pythonusb\\pythonusb\\python\\python-3.9.9\\lib\\site-packages (from pandas>=0.24.2->mlxtend) (2021.3)\n",
      "Requirement already satisfied: threadpoolctl>=2.0.0 in e:\\pythonusb\\pythonusb\\python\\python-3.9.9\\lib\\site-packages (from scikit-learn>=0.20.3->mlxtend) (3.1.0)\n",
      "Requirement already satisfied: six>=1.5 in e:\\pythonusb\\pythonusb\\python\\python-3.9.9\\lib\\site-packages (from python-dateutil>=2.7->matplotlib>=3.0.0->mlxtend) (1.16.0)\n",
      "Requirement already satisfied: tomli>=1.0.0 in e:\\pythonusb\\pythonusb\\python\\python-3.9.9\\lib\\site-packages (from setuptools-scm>=4->matplotlib>=3.0.0->mlxtend) (1.2.2)\n",
      "Note: you may need to restart the kernel to use updated packages.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING: You are using pip version 22.0.4; however, version 22.1 is available.\n",
      "You should consider upgrading via the 'E:\\PythonUSB\\PythonUSB\\Python\\python-3.9.9\\pythonw.exe -m pip install --upgrade pip' command.\n"
     ]
    }
   ],
   "source": [
    "pip install mlxtend"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9e3b6ae8-a3c6-4fd5-9c47-8612600d6b01",
   "metadata": {},
   "source": [
    "Mlxtendの情報を、pip show コマンドで見ます。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "be151e43-e34b-4c65-ad00-4228259fbc47",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Name: mlxtend\n",
      "Version: 0.19.0\n",
      "Summary: Machine Learning Library Extensions\n",
      "Home-page: https://github.com/rasbt/mlxtend\n",
      "Author: Sebastian Raschka\n",
      "Author-email: mail@sebastianraschka.com\n",
      "License: BSD 3-Clause\n",
      "Location: e:\\pythonusb\\pythonusb\\python\\python-3.9.9\\lib\\site-packages\n",
      "Requires: joblib, matplotlib, numpy, pandas, scikit-learn, scipy, setuptools\n",
      "Required-by: \n",
      "Note: you may need to restart the kernel to use updated packages.\n"
     ]
    }
   ],
   "source": [
    "pip show mlxtend"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "239ba1a4-3a87-4a15-b25c-1c37c4efe51b",
   "metadata": {},
   "source": [
    "再度、mlxtend.plotting.scatterplotmatrixをインポートします。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "3bba2b48-6bf6-4876-b781-8cc25db35267",
   "metadata": {},
   "outputs": [],
   "source": [
    "from mlxtend.plotting import scatterplotmatrix"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "78a462f8-6cf1-487e-b416-dd7c14400bc5",
   "metadata": {},
   "source": [
    "散布図マトリクス（scatterplot matrix）を描きます。テキトーに使うと下のようにエラーが生じます。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "dbbc8b31-f4c4-421a-919b-41e7d8ac6cc6",
   "metadata": {},
   "outputs": [
    {
     "ename": "InvalidIndexError",
     "evalue": "(slice(None, None, None), 1)",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mTypeError\u001b[0m                                 Traceback (most recent call last)",
      "\u001b[1;32mE:\\PythonUSB\\PythonUSB\\Python\\python-3.9.9\\lib\\site-packages\\pandas\\core\\indexes\\base.py\u001b[0m in \u001b[0;36mget_loc\u001b[1;34m(self, key, method, tolerance)\u001b[0m\n\u001b[0;32m   3620\u001b[0m             \u001b[1;32mtry\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m-> 3621\u001b[1;33m                 \u001b[1;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_engine\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mget_loc\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mcasted_key\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m   3622\u001b[0m             \u001b[1;32mexcept\u001b[0m \u001b[0mKeyError\u001b[0m \u001b[1;32mas\u001b[0m \u001b[0merr\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32mE:\\PythonUSB\\PythonUSB\\Python\\python-3.9.9\\lib\\site-packages\\pandas\\_libs\\index.pyx\u001b[0m in \u001b[0;36mpandas._libs.index.IndexEngine.get_loc\u001b[1;34m()\u001b[0m\n",
      "\u001b[1;32mE:\\PythonUSB\\PythonUSB\\Python\\python-3.9.9\\lib\\site-packages\\pandas\\_libs\\index.pyx\u001b[0m in \u001b[0;36mpandas._libs.index.IndexEngine.get_loc\u001b[1;34m()\u001b[0m\n",
      "\u001b[1;31mTypeError\u001b[0m: '(slice(None, None, None), 1)' is an invalid key",
      "\nDuring handling of the above exception, another exception occurred:\n",
      "\u001b[1;31mInvalidIndexError\u001b[0m                         Traceback (most recent call last)",
      "\u001b[1;32m~\\AppData\\Local\\Temp/ipykernel_15172/3459118802.py\u001b[0m in \u001b[0;36m<module>\u001b[1;34m\u001b[0m\n\u001b[1;32m----> 1\u001b[1;33m \u001b[0mscatterplotmatrix\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mdf_room\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mfigsize\u001b[0m\u001b[1;33m=\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;36m10\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;36m8\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m",
      "\u001b[1;32mE:\\PythonUSB\\PythonUSB\\Python\\python-3.9.9\\lib\\site-packages\\mlxtend\\plotting\\scatterplotmatrix.py\u001b[0m in \u001b[0;36mscatterplotmatrix\u001b[1;34m(X, fig_axes, names, figsize, alpha, **kwargs)\u001b[0m\n\u001b[0;32m     72\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     73\u001b[0m     \u001b[1;32mfor\u001b[0m \u001b[0mi\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mj\u001b[0m \u001b[1;32min\u001b[0m \u001b[0mzip\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m*\u001b[0m\u001b[0mnp\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mtriu_indices_from\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0maxes\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mk\u001b[0m\u001b[1;33m=\u001b[0m\u001b[1;36m1\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m---> 74\u001b[1;33m         \u001b[0maxes\u001b[0m\u001b[1;33m[\u001b[0m\u001b[0mj\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mi\u001b[0m\u001b[1;33m]\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mscatter\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mX\u001b[0m\u001b[1;33m[\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mj\u001b[0m\u001b[1;33m]\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mX\u001b[0m\u001b[1;33m[\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mi\u001b[0m\u001b[1;33m]\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0malpha\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0malpha\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;33m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m     75\u001b[0m         \u001b[0maxes\u001b[0m\u001b[1;33m[\u001b[0m\u001b[0mj\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mi\u001b[0m\u001b[1;33m]\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mset_xlabel\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mnames\u001b[0m\u001b[1;33m[\u001b[0m\u001b[0mj\u001b[0m\u001b[1;33m]\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     76\u001b[0m         \u001b[0maxes\u001b[0m\u001b[1;33m[\u001b[0m\u001b[0mj\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mi\u001b[0m\u001b[1;33m]\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mset_ylabel\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mnames\u001b[0m\u001b[1;33m[\u001b[0m\u001b[0mi\u001b[0m\u001b[1;33m]\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32mE:\\PythonUSB\\PythonUSB\\Python\\python-3.9.9\\lib\\site-packages\\pandas\\core\\frame.py\u001b[0m in \u001b[0;36m__getitem__\u001b[1;34m(self, key)\u001b[0m\n\u001b[0;32m   3503\u001b[0m             \u001b[1;32mif\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mcolumns\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mnlevels\u001b[0m \u001b[1;33m>\u001b[0m \u001b[1;36m1\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   3504\u001b[0m                 \u001b[1;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_getitem_multilevel\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mkey\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m-> 3505\u001b[1;33m             \u001b[0mindexer\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mcolumns\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mget_loc\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mkey\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m   3506\u001b[0m             \u001b[1;32mif\u001b[0m \u001b[0mis_integer\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mindexer\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   3507\u001b[0m                 \u001b[0mindexer\u001b[0m \u001b[1;33m=\u001b[0m \u001b[1;33m[\u001b[0m\u001b[0mindexer\u001b[0m\u001b[1;33m]\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32mE:\\PythonUSB\\PythonUSB\\Python\\python-3.9.9\\lib\\site-packages\\pandas\\core\\indexes\\base.py\u001b[0m in \u001b[0;36mget_loc\u001b[1;34m(self, key, method, tolerance)\u001b[0m\n\u001b[0;32m   3626\u001b[0m                 \u001b[1;31m#  InvalidIndexError. Otherwise we fall through and re-raise\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   3627\u001b[0m                 \u001b[1;31m#  the TypeError.\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m-> 3628\u001b[1;33m                 \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_check_indexing_error\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mkey\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m   3629\u001b[0m                 \u001b[1;32mraise\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   3630\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32mE:\\PythonUSB\\PythonUSB\\Python\\python-3.9.9\\lib\\site-packages\\pandas\\core\\indexes\\base.py\u001b[0m in \u001b[0;36m_check_indexing_error\u001b[1;34m(self, key)\u001b[0m\n\u001b[0;32m   5635\u001b[0m             \u001b[1;31m# if key is not a scalar, directly raise an error (the code below\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   5636\u001b[0m             \u001b[1;31m# would convert to numpy arrays and raise later any way) - GH29926\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m-> 5637\u001b[1;33m             \u001b[1;32mraise\u001b[0m \u001b[0mInvalidIndexError\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mkey\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m   5638\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   5639\u001b[0m     \u001b[1;33m@\u001b[0m\u001b[0mcache_readonly\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;31mInvalidIndexError\u001b[0m: (slice(None, None, None), 1)"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x576 with 36 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "scatterplotmatrix(df_room, figsize=(10, 8))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "55f06b4c-af7c-4684-97e2-8ab6c156c987",
   "metadata": {},
   "source": [
    "スカラーだけ使えとのことですので、スカラーのカラムだけ取り出します。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "882bd584-44f5-4fea-b018-bf9fd9520112",
   "metadata": {},
   "outputs": [],
   "source": [
    "cols = ['rent','size','age']\n",
    "X = df_room[cols]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "75e6608f-af40-462f-b361-71b8dacc667b",
   "metadata": {},
   "source": [
    "これらの列どうしの散布図マトリクスを描きます。values属性で値だけ取り出さないと描けないことに注意しましょう。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "950addaf-1ba0-496e-a1e6-521a111bb208",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(<Figure size 576x576 with 9 Axes>,\n",
       " array([[<AxesSubplot:xlabel='X1', ylabel='Count'>, <AxesSubplot:>,\n",
       "         <AxesSubplot:>],\n",
       "        [<AxesSubplot:xlabel='X2', ylabel='X1'>,\n",
       "         <AxesSubplot:xlabel='X2', ylabel='Count'>, <AxesSubplot:>],\n",
       "        [<AxesSubplot:xlabel='X3', ylabel='X1'>,\n",
       "         <AxesSubplot:xlabel='X3', ylabel='X2'>,\n",
       "         <AxesSubplot:xlabel='X3', ylabel='Count'>]], dtype=object))"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x576 with 9 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "scatterplotmatrix(X.values)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "af6d8cd5-9abd-453b-b0d6-4a40efe83a28",
   "metadata": {},
   "source": [
    "一方で、**Seaborn**を使えば、データフレームをそのまま使えます。カテゴリー変数を除く必要はありません。自動的に除いて処理されます。"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2fd234bc-ec0b-4885-b877-3b6b8f1195f9",
   "metadata": {},
   "source": [
    "Seabornはsnsという名前でインポートするのが慣例です。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "71649e70-7e4d-4ddb-8c86-c5b369ee4e73",
   "metadata": {},
   "outputs": [],
   "source": [
    "import seaborn as sns"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1a7bbdab-c5b2-4b74-8ad0-5b20bd7da808",
   "metadata": {},
   "source": [
    "Seabornのpairplot()関数を用いて散布図を描きます。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "51223c03-a073-4212-bdcb-eecf48a9e6c7",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<seaborn.axisgrid.PairGrid at 0x227d6b6d070>"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 540x540 with 12 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.pairplot(df_room)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a5cf9e4b-cefe-4f4a-90c3-49f82ed31e94",
   "metadata": {},
   "source": [
    "駅からの近さで色分けしてみます。散布図をpairplot()関数で書くときに、どのカテゴリ変数で色分けしたいかをhue属性で指定します。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "132d1423-88af-4cad-85af-b4f5fca92b65",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<seaborn.axisgrid.PairGrid at 0x227d55346a0>"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 594.875x540 with 12 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.pairplot(df_room, hue='proximity')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1b86150c-e3dd-4207-8ad3-0c9dd79d8c20",
   "metadata": {},
   "source": [
    "間取りで色分けしてみます。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "9fa4b14f-3232-4989-b23b-fc86d64ce661",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<seaborn.axisgrid.PairGrid at 0x227d5f4dc40>"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 608.375x540 with 12 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.pairplot(df_room, hue='plan')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "288779f0-6e69-491a-b59e-60697d90a196",
   "metadata": {},
   "source": [
    "部屋の向きで色分けします。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "bda4764d-8bb1-4ffa-99b7-46748264f69c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<seaborn.axisgrid.PairGrid at 0x227d5f4db20>"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 593.375x540 with 12 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.pairplot(df_room, hue='direction')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "86686d89-ad84-48a4-bcb7-30a821c55d29",
   "metadata": {},
   "source": []
  },
  {
   "cell_type": "markdown",
   "id": "8ad7c17e-276d-4aee-a21c-15f032607e60",
   "metadata": {
    "tags": []
   },
   "source": [
    "### 共分散"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3c3f34d1-328a-4b27-9eb4-29d77d06a355",
   "metadata": {},
   "source": [
    "相関の強さを数値化する方法のひとつに、相関係数（ピアソンの相関係数）があります。それを計算するためには、「**共分散**」を計算する必要があります。"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0b31bc11-d831-4361-a74a-aadd0d9ddaf7",
   "metadata": {},
   "source": [
    "変数$x$と$y$のペア$(x_i, y_i)$が$(x_1,y_1)$から$(x_n, y_n)$まで$n$個あるとします。次に示すような形で計算できる**共分散**$s_{xy}$は、変数$x$と$y$がそれぞれの平均$\\bar{x}$, $\\bar{y}$から互いに同じ方向あるいは正反対の方向に変動することが多いほど絶対値が大きくなります。同じ方向に動く傾向があれば正の値に、反対方向に動く傾向があれば、負の値に、それぞれ、絶対値が大きくなります。"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cf6daa12-fef0-48cb-b46e-3248f2e1965e",
   "metadata": {},
   "source": [
    "$$s_{xy} = \\frac{1}{n}\\{(x_1 - \\bar{x})(y_1 - \\bar{y})+(x_2 - \\bar{x})(y_2 - \\bar{y})+ ... +(x_n - \\bar{x})(y_n - \\bar{y})\\}$$ $$= \\frac{1}{n}\\sum_{i=1}^n(x_i - \\bar{x})(y_i - \\bar{y})$$"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e6aeadf2-867f-41ce-b3c0-6a8a6468e123",
   "metadata": {},
   "source": [
    "そして、$x$と$y$とが**独立**ー互いにてんでばらばらの方向に動くーの場合$s_{xy}$は$0$になります。"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5c2b9498-97b5-4010-8361-6786c20ded1c",
   "metadata": {},
   "source": [
    "### 相関係数"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "68524c77-b3fc-43aa-92bd-797bee10d740",
   "metadata": {},
   "source": [
    "**共分散**$s_{xy}$を、$x$の**標準偏差**$s_x$と$y$の**標準偏差**$s_y$で割ることで、この値は$-1$と$1$との間になり、正規化されます。これを**相関係数**（**ピアソン**の相関係数）$r_{xy}$といいます。"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2debd7a1-1576-415a-b255-c222178c2fe3",
   "metadata": {},
   "source": [
    "$$r_{xy} = \\frac{s_{xy}}{s_x s_y} = \\frac{1}{n}\\sum_{i=1}^n\\frac{(x_i - \\bar{x})}{s_x}\\frac{(y_i - \\bar{y})}{s_y}$$"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ef5067c8-367a-4dd2-abff-7948d0e0f241",
   "metadata": {},
   "source": [
    "そして$s_x$も$s_y$も分母が$\\sqrt{n}$ですから、\n",
    "$$r_{xy}= \\frac{\\sum_{i=1}^n(x_i - \\bar{x})(y_i - \\bar{y})}{\\sqrt{\\sum_{i=1}^n(x_i - \\bar{x})^2}\\sqrt{\\sum_{i=1}^n(y_i - \\bar{y})^2}}$$"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0ce0560c-de4e-446d-89cd-fa790de24c58",
   "metadata": {},
   "source": [
    "$x$と$y$が同じ方向にいっしょに動く傾向が強ければ$r_{xy}$は$1$に近くなり、反対方向に動く傾向が強ければ、$r_{xy}$は$-1$に近くなります。"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a19a7404-e8ff-4f06-8104-ea1333004ce9",
   "metadata": {},
   "source": [
    "それでは、家賃、広さ、築年数の間の相関係数を求めましょう。ペアは３種類ありますから、相関係数も３個できます。  \n",
    "相関係数は、Pandasデータフレームの.corr()メソッドで求めます。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "c96b6afa-5352-4495-8448-8ca69839faaf",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "          rent      size       age\n",
      "rent  1.000000  0.841193  0.245452\n",
      "size  0.841193  1.000000  0.515945\n",
      "age   0.245452  0.515945  1.000000\n"
     ]
    }
   ],
   "source": [
    "corr_room = df_room.corr()\n",
    "print(corr_room)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e9edd0df-1e4f-4ba3-974c-8ce52b2cb47c",
   "metadata": {},
   "source": [
    "Seabornのheatmap()関数で「ヒートマップ」として表示しましょう。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "id": "d3afdf2c-d80e-43c2-81e7-0a239d48c1de",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:>"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.heatmap(corr_room, cmap='coolwarm', vmin=-1, vmax=1, annot=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "814e4baf-76b0-4026-801e-efc0b770a2b1",
   "metadata": {},
   "source": [
    "これからわかるように、これら３つの変数の間にはどれの間にも正の相関があります。そしてもっとも強いのは広さと家賃との関係です。"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f66f3da1-ffbf-4c54-8887-a13470293809",
   "metadata": {},
   "source": [
    "気になるのは、築年数と家賃との間に正の相関があることです。古いほど、家賃が高くなる傾向です。これは、直観と反しますね。   \n",
    "その関係から、広さの影響を取り除く必要があります。上の表からわかるように、築年数と広さとの間には正の相関があるからです。つまり、古い→広い→高い、という関係が含まれているかもしれません。"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f7fba9c0-010e-46a7-b617-665f4a1079cc",
   "metadata": {},
   "source": [
    "家賃と広さとの相関係数を$r_{xy}$、家賃と築年数との相関係数を$r_{xz}$、広さと築年数との相関係数を$r_{yz}$とします。広さの影響を除いた、築年数と家賃との相関係数$r_{xz-y}$を、計算するには、次のようにします。"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0c4840e8-301f-4250-9968-bba17403933a",
   "metadata": {},
   "source": [
    "$$r_{xz-y} = \\frac{r_{xz} - r_{xy}r_{yz}}{\\sqrt{1 - r_{xy}^2}\\sqrt{1 - r_{yz}^2}}$$"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "dfb5eb34-0a7b-4510-b6a7-23c7e9affe9c",
   "metadata": {},
   "source": [
    "この相関係数$r_{xz-y}$を**偏相関係数**といいます。  "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8c9b7616-0130-4956-9c3a-c29c6153bb6a",
   "metadata": {},
   "source": [
    "実際に計算してみましょう。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "80e502e4-ba13-47e1-9f47-c1eb7bb91572",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.8411926135904368\n"
     ]
    }
   ],
   "source": [
    "rxy = corr_room.loc[\"rent\",\"size\"]\n",
    "print(rxy)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "id": "18cf872d-413c-4cda-8181-9bdd3a440aeb",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.24545151007620591\n"
     ]
    }
   ],
   "source": [
    "rxz = corr_room.loc[\"rent\",\"age\"]\n",
    "print(rxz)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "id": "3b5c3ecc-c796-4d1e-8385-a720cd0513fb",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.5159447489463378\n"
     ]
    }
   ],
   "source": [
    "ryz = corr_room.loc[\"size\",\"age\"]\n",
    "print(ryz)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "id": "3fe5de13-750a-4654-b2de-bca48fdbc22c",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "-0.4070703491024938\n"
     ]
    }
   ],
   "source": [
    "rxz_y = (rxz - rxy*ryz)/(np.sqrt(1 - rxy**2)*np.sqrt(1 - ryz**2))\n",
    "print(rxz_y)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "31a52e05-f0b8-473e-90ff-168e5deb5e4e",
   "metadata": {},
   "source": [
    "このように、広さの影響を除いたあとの、家賃と築年数との偏相関係数は負になりました。同じ大きさの部屋であるなら、古い物件ほど家賃が安いことになります。"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "770a36ea-3501-41cf-bfb5-a96b9b3f500c",
   "metadata": {},
   "source": [
    "#### 例題"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "937aeef3-f434-4508-b604-3ef5d9c2962c",
   "metadata": {},
   "source": [
    "次の表は、10人の身長・体重の観察値です。身長と体重の関係の散布図を表示し、相関係数を求めましょう。"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0299c8b9-e99f-43f7-9841-1be5feda3fa4",
   "metadata": {},
   "source": [
    "|身長|158|167|150|169|160|159|152|162|156|164|\n",
    "|--|--|--|--|--|--|--|--|--|--|--|\n",
    "|体重|55|57|45|64|50|60|53|49|44|57|"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "37851d9b-c9cf-497d-b87f-d4a7526b5523",
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "%matplotlib inline\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "24c1d01d-e547-4741-ab3a-ee4335d3a7b1",
   "metadata": {},
   "source": [
    "データは次の通りです。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "b73b5d6f-28da-4377-a672-8c78db55f76c",
   "metadata": {},
   "outputs": [],
   "source": [
    "height = np.array([158,167,150,169,160,159,152,162,156,164])\n",
    "weight = np.array([55,57,45,64,50,60,53,49,44,57])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6d8eb451-1cf6-4b28-9496-002c69d9da09",
   "metadata": {},
   "source": [
    "データフレームをつくります。関数pandas.DataFrame()の引数が辞書型であることに注意してください。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "d50fecd7-b65e-4f6c-b8cf-be2cb2b61b83",
   "metadata": {},
   "outputs": [],
   "source": [
    "df_hw = pd.DataFrame({'Height':height,'Weight':weight})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "d1b01040-bc2e-494c-821e-0f277e772fde",
   "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>Height</th>\n",
       "      <th>Weight</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>158</td>\n",
       "      <td>55</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>167</td>\n",
       "      <td>57</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>150</td>\n",
       "      <td>45</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>169</td>\n",
       "      <td>64</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>160</td>\n",
       "      <td>50</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>159</td>\n",
       "      <td>60</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>152</td>\n",
       "      <td>53</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>162</td>\n",
       "      <td>49</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>156</td>\n",
       "      <td>44</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>164</td>\n",
       "      <td>57</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Height  Weight\n",
       "0     158      55\n",
       "1     167      57\n",
       "2     150      45\n",
       "3     169      64\n",
       "4     160      50\n",
       "5     159      60\n",
       "6     152      53\n",
       "7     162      49\n",
       "8     156      44\n",
       "9     164      57"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_hw"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b38ffb0a-59bd-4890-bb74-093286f12131",
   "metadata": {},
   "source": [
    "散布図を描きましょう。まず、Seabornでペアプロットとして描きましょう。これの利点は各変数単独での分布もわかることです。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "7b20885e-d9c6-43f4-85c4-f4107b293c74",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<seaborn.axisgrid.PairGrid at 0x2da9ef367c0>"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 360x360 with 6 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.pairplot(df_hw)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "bca27053-3dff-4ead-ae6e-3390c93c2fa1",
   "metadata": {},
   "source": [
    "単純に、Matplotlibで描いてみましょう。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "624f8fec-7967-4ea9-a946-441daf83745d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.collections.PathCollection at 0x2da9f3a25b0>"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.scatter(height,weight)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0ee6015d-6fde-42d3-b81d-d49811350179",
   "metadata": {},
   "source": [
    "Pandaデータフレームのplot()メソッドでも描けます。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "5445047e-31a3-4e58-b2e2-ae52e010bbb0",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='Height', ylabel='Weight'>"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "df_hw.plot('Height','Weight',kind='scatter')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c6e2e45a-4880-408b-807f-ca53a6e4c85a",
   "metadata": {},
   "source": [
    "相関係数を求めましょう。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "6b8c7635-fa2a-4df6-b3fd-101b86bd03e5",
   "metadata": {},
   "outputs": [],
   "source": [
    "corr_hw = df_hw.corr()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "e31e7d4b-1fb9-4f7b-b389-d95c70a7a1da",
   "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>Height</th>\n",
       "      <th>Weight</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Height</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.681979</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Weight</th>\n",
       "      <td>0.681979</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          Height    Weight\n",
       "Height  1.000000  0.681979\n",
       "Weight  0.681979  1.000000"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "corr_hw"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7329dc9a-a298-41f0-8163-855516c8f082",
   "metadata": {},
   "source": [
    "上の表より、0.682とわかります。"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0f680292-9c55-49d5-8bb6-b410ecff84f1",
   "metadata": {},
   "source": [
    "上の散布図を見ると、直線関係が見えてくると思います。それを定量的に示すのが、次に取扱う回帰分析です。"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "bc68d3c1-d2f3-4fb6-a06c-cf852b885232",
   "metadata": {},
   "source": [
    "## 回帰分析"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0a1a11d7-2cc9-4557-bd0a-f8e0ef85f3ab",
   "metadata": {},
   "source": [
    "### 独立変数と従属変数"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "44a211db-c5d2-416b-9605-2c55fc08069b",
   "metadata": {},
   "source": [
    "回帰分析は、ある変数xから別の変数yの値を予測するために行います。前提として、xとyは対等ではなく、xを原因としてyが起こるというような、方向性のある関係を想定しています。"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e1483b76-da08-4993-9ee7-fc30020bf9f5",
   "metadata": {},
   "source": [
    "例えば部屋の広さ$x$[$m^2$]で家賃$y$[円]が説明できると考えます。このとき、説明するほうの変数$x$を説明変数（独立変数、予測変数などともいう）、説明されるほうの変数$y$を応答変数（目的変数、従属変数、被説明変数などともいう）といいます。"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e840ec85-7a42-44be-8ae0-5301d1393fab",
   "metadata": {},
   "source": [
    "### 回帰直線"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "95dec667-56bd-4703-96cd-4d3de2303c63",
   "metadata": {},
   "source": [
    "部屋の大きさを$x$, 家賃を$y$とおき、その間に$$y = \\alpha + \\beta x$$という直線関係を考え、部屋の大きさ$x$から家賃$y$を予測します。この直線を$y$の$x$に対する回帰直線（regression line）とよび、$\\alpha$と$\\beta$を回帰係数（regression coefficient）とよびます。  \n",
    "この回帰直線を散布図のデータと対応させ、最もよくあてはまる回帰係数を推定します。"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a0b73ad3-98b9-4996-bcf7-46969533ce31",
   "metadata": {},
   "source": [
    "#### 最小二乗法"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "bef11103-5198-492e-b9bb-75d5c1180797",
   "metadata": {},
   "source": [
    "そのような回帰直線はどのようにして求めるのでしょうか。仮に$\\alpha$, $\\beta$をある値$\\hat{\\alpha}$, $\\hat{\\beta}$にしたとき、$x_i$に対する$y_i$の**予測値**$\\hat{y}_i$は、$\\hat{y}_i = \\hat{\\alpha} + \\hat{\\beta}x_i$となります。"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5439712a-9f25-4e90-8697-b45c11ba4a47",
   "metadata": {},
   "source": [
    "観測値の実際の値$y_i$と回帰直線で予測した値$\\hat{y}_i$との差$e_i = y_i - \\hat{y}_i$を**残差residual**と呼びます。つまり予測誤差です。"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "84d3bfc7-83a1-411b-b558-ecfecd0d788a",
   "metadata": {},
   "source": [
    "観測値は$n$個ですから、残差の二乗和（残差平方和）は、$\\hat{\\alpha}$と$\\hat{\\beta}$の関数として、"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c9c461aa-a9ff-4ad3-aa25-c289d87419d9",
   "metadata": {},
   "source": [
    "$$S_{(\\hat{\\alpha},\\hat{\\beta})} = \\sum_{i = 1}^n e_i^2 = \\sum_{i = 1}^n (y_i - \\hat{y}_i)^2-  \n",
    "= \\sum_{i = 1}^n\\{y_i - (\\hat{\\alpha} + \\hat{\\beta} x_i) \\}^2$$"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4e2ac08e-c0e6-4c99-8b94-144ce35ea24f",
   "metadata": {},
   "source": [
    "のように表され、これを最小にするような$\\hat{\\alpha}$と$\\hat{\\beta}$を求めればよいことになります。"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "aeb6e841-15b0-4b7b-8ad1-62f03c0c333c",
   "metadata": {},
   "source": [
    "$S_{(\\hat{\\alpha},\\hat{\\beta})}$は２つのパラメータ$\\hat{\\alpha}$と$\\hat{\\beta}$の２次関数で、最小値が存在しますから、$\\hat{\\alpha}$と$\\hat{\\beta}$で偏微分し$0$に等しいとすると、次の２つの方程式ができます。"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "63b0d877-ae00-4bce-80b1-e0189fb990cd",
   "metadata": {},
   "source": [
    "$$n\\hat{\\alpha} + \\hat{\\beta}\\sum x_i = \\sum y_i$$"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c4b7de8d-0c55-45a1-ab62-84eb4fb9a925",
   "metadata": {},
   "source": [
    "$$\\hat{\\alpha}\\sum x_i + \\hat{\\beta}\\sum x_i^2 = \\sum x_i y_i$$"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2ee2302a-d749-45ea-a393-19b7787579c1",
   "metadata": {},
   "source": [
    "これらは、$\\hat{\\alpha}$と$\\hat{\\beta}$の連立１次方程式です。そして、"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9d4ac35c-c5e8-4285-aa81-853517d28cac",
   "metadata": {},
   "source": [
    "$$\\hat{\\beta} = \\frac{\\sum(y_i - \\bar{y})(x_i - \\bar{x})}{\\sum (x_i - \\bar{x})^2} = \\frac{s_{xy}}{s_{xx}} = r_{xy}\\frac{\\sqrt{s_{yy}}}{\\sqrt{s_{xx}}} = r_{xy}\\frac{s_y}{s_x}$$"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8682723c-ddf5-4264-aaba-588957d6c81b",
   "metadata": {},
   "source": [
    "$$\\hat{\\alpha} = \\bar{y} - \\hat{\\beta}\\bar{x}$$"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4072b1d0-fb0a-408a-8889-a53dd1ec8363",
   "metadata": {},
   "source": [
    "となります。なお、$s_{xx}$, $s_{yy}$はそれぞれ$x$, $y$の分散を意味します。同様に、$s_{xy}$は$x$と$y$の共分散です。"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d88d3837-76e8-4dc1-b99f-01ec33c70735",
   "metadata": {},
   "source": [
    "また、$s_x$, $s_y$は$x$, $y$の標準偏差です。"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7669ff0a-5113-47e3-87fb-2c6962820bb8",
   "metadata": {},
   "source": [
    "上の式から、次のような関係がわかります。\n",
    "- 回帰係数$\\hat{\\beta}$は、$x$, $y$の共分散と$x$の分散の比である。\n",
    "- 回帰係数$\\hat{\\beta}$は同時に、$x$, $y$の相関係数に$y$と$x$の標準偏差の比をかけたものでもある。"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e7e3f97a-de7e-43b2-aa48-1908f469aeb3",
   "metadata": {},
   "source": [
    "このような**最小二乗法**による$\\hat{\\alpha}$, $\\hat{\\beta}$の解には次のような特徴があります。"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0a1bbc70-dca1-40ef-8750-c3846637778c",
   "metadata": {},
   "source": [
    "- 予測値$\\hat{y}_i = \\hat{\\alpha} + \\hat{\\beta} x_i$の平均は観測値$y_i$の平均と等しい。\n",
    "- 残差$e_i = y_i - \\hat{y}_i$の平均は$0$となる。\n",
    "- 回帰直線は点$(\\bar{x}, \\bar{y})$を通る。\n",
    "- 予測値$\\hat{y}_i$と残差$e_i$との相関係数は0である。\n",
    "- $y$の**変動の大きさ**を表す**平方和$S_y$は、回帰による平方和$S_R$と残差平方和$S_e$とに分解できる**。つまり、$S_y = S_R + S_e$"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "57afc8d2-2878-4700-83e4-92377368a666",
   "metadata": {},
   "source": [
    "そして、$y$の平方和において回帰による平方和の割合、すなわち$$R^2 = \\frac{S_R}{S_y}$$を、**決定係数**といいます。これが大きい（ただし最大値は1）ほど、$y$の変動のうち回帰直線で説明できる部分の割合が大きくなります。"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "69aac0b2-e2b1-4aa7-b9fd-b2b70459b500",
   "metadata": {},
   "source": [
    "#### 回帰モデルの推定と検定"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c849e362-251e-42f6-bd06-855d7e216b7b",
   "metadata": {},
   "source": [
    "- $F$統計量：$S_R$と$S_e$を比較する指標です。帰無仮説$H_0$を、「すべての従属変数の回帰係数が0」としてF統計量を計算します。$F$統計量が大きいほど、その実現確率は小さくなります。\n",
    "- $F$統計量の確率：$H_0$のもとで、計算されたような値を得る確率です。有意水準と比較して回帰の優位性の検定に用います。\n",
    "- $t$統計量：$F$統計量のもとで、回帰が有意であると判断されたとき、個別の回帰係数の信頼区間を求めたり有意性の検定をしたりするために用います。回帰係数が0であるという帰無仮説のもとで計算された$\\hat{\\beta}$の値から計算します。大きいほど、0から外れます。\n",
    "- $P-$値：上記帰無仮説のもとで、$t$統計量がその値よりも極端な値をとる確率です。"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "719516a9-fd6b-432c-98a7-0e18f77de03d",
   "metadata": {},
   "source": [
    "### Statsmodelsを使った回帰分析（単回帰分析）"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c077587a-de3c-4fe3-b33a-dfffbfb39c94",
   "metadata": {},
   "source": [
    "それでは、家賃データから、広さと家賃との関係を表す回帰直線を求めましょう。"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d85a6769-92da-45d4-9391-f826510c833d",
   "metadata": {},
   "source": [
    "statsmodelsのモジュールをインポートします。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "1a6a147e-475c-454d-99c8-d5bb055fe7f9",
   "metadata": {},
   "outputs": [],
   "source": [
    "import statsmodels.formula.api as smf\n",
    "import statsmodels.api as sm"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4361dd99-1dd6-483a-8fe0-18ffb5761edb",
   "metadata": {},
   "source": [
    "データは、すでに読み込んである、df_roomという名前のデータフレームです。  \n",
    "説明変数と目的変数を指定します。説明変数(x)はsize, 目的変数(y)はrentです。"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "59964dab-1cea-42cc-bc9a-897e08113ef6",
   "metadata": {},
   "source": [
    "その前に関係を確認しておきます。散布図は、Pandasデータフレームのplot()メソッドで描けます。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "e40731d8-6043-4e1a-9415-5bda26a57e70",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='size', ylabel='rent'>"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAZcAAAEGCAYAAACpXNjrAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8/fFQqAAAACXBIWXMAAAsTAAALEwEAmpwYAAAgr0lEQVR4nO3de3BcZ5nn8e8jyVIcSyBFNgIsCYWRByrkIoIK7DJFsQkEw1IxW4RsuJTMbpbU7rC1zALrBKaKzHqWqgkwZGCXSW0mZIiHWUImGbCHBTKeJCyzs0lAAcVJuEXkJjnBMY4UfJV1efaPfjvu7nQftVrdp/uc/n2qZKvfc07320eWn37eq7k7IiIi1dRS7wqIiEj6KLiIiEjVKbiIiEjVKbiIiEjVKbiIiEjVtdW7Ao1i/fr1PjQ0VO9qiIgkygMPPPBbd99QWK7gEgwNDTE+Pl7vaoiIJIqZPVmsXM1iIiJSdQouIiJSdQouIiJSdQouIiJSdQouIiJSdQouIk3s8NE5Hpya5fDRuXpXRVJGQ5FFmtSeiQNcfcd+1rS0ML+0xOfeez6Xjmysd7UkJZS5iDShw0fnuPqO/ZycX+LI3AIn55fYecd+ZTBSNQouIk1oeuYEa1ryf/3XtLQwPXOiTjWStFFwEWlC/T1rmV9ayiubX1qiv2dtnWokaaPgItKEejs7+Nx7z+eMNS10dbRxxpoWPvfe8+nt7HjhnMmDR7h9fIrJg0fqWFNJKnXoizSpS0c2snV4PdMzJ+jvWZsXWD7z7YfYfd9TLzwe2zLIru3n1aOaklDKXESaWG9nBxcMdL8oY8kNLAC7731KGYysiIKLiOSZmJpdUblIMQouIpJnZKB7ReUixSi4iEie4b4uxrYM5pWNbRlkuK+rTjWSJFKHvoi8yK7t5zG2eYiJqVlGBroVWGTFFFxEUu7w0bmiI8KWM9zXVdWgUmk9kqwZ33OWgotIijXK+mGNUo84NeN7zqU+F5GUapT1wxqlHnFqxvdcSMFFJKUaZf2wRqlHnJrxPRdScBFJqUZZP6xR6hGnZnzPhRRcRFKqnPXDmqkecWrG91zI3L3edWgIo6OjPj4+Xu9qiFRdo4xYapR6xKkZ3rOZPeDuo4XlGi0mknK9nR0N8R9bo9QjTs34nrPULCappL3hRepLmYukTrPPLxBpBMpcJFU0v0CkMSi4SKpofoFIY6hZcDGzm83sWTN7uMixT5iZm9n68NjM7MtmNmlm+83swpxzd5jZo+FrR075G8zsoXDNl83MQvlZZrYvnL/PzHpq9R6l8Wh+gUhjqGXm8jVgW2GhmQ0AlwC5W929E9gUvq4CbgjnngVcC7wJeCNwbU6wuAH4SM512de6BrjL3TcBd4XH0iQ0v0CkMdSsQ9/df2hmQ0UOXQ/sBPbklG0Hdntm0s19ZtZtZq8A3grsc/fnAMxsH7DNzH4AvMTd7wvlu4H3AN8Lz/XW8Ly3AD8Arq7iW5MGF7U3vIjEI9bRYma2HTjg7g+GVqysjcBUzuPpUBZVPl2kHKDP3Z8J3/8G6Iuoz1VkMiUGBwdLnSYJ1MzzC+qpGSYNpk2tfmaxBRczOxP4NJkmsVi4u5tZySUI3P1G4EbIzNCPq14iaaQh4MlTy59ZnKPFfg84G3jQzJ4A+oGfmNnLgQPAQM65/aEsqry/SDnAwdCkRvj72aq/ExHJoyHgyVPrn1lswcXdH3L3l7n7kLsPkWnKutDdfwPsBcbCqLHNwPOhaetO4BIz6wkd+ZcAd4ZjvzOzzWGU2Bin+3D2AtlRZTvI79sRkRrQEPDkqfXPrJZDkb8B3Au8xsymzezKiNO/CzwGTAJ/CfwBQOjI/xPgx+FrV7ZzP5xzU7jm12Q68wH+FHi7mT0KvC08FpEa0hDw5Kn1z0yrIgdaFVlkdfZOHGCn+lwSpRo/s1KrIiu4BAouIqun0WLJs9qfmZbcF5Ga0xDw5KnVz0xri4mISNUpuIgkyOTBI9w+PsXkwSM1f60498TR/jvpo2YxkYT4zLcfYvd9p5fkG9syyK7t59XkteKcEKnJl+mkzEUkASYPHskLLAC7732qJhlMnBMiNfkyvRRcRBJgYmp2ReWrEeeESE2+TC8FF5EEGBnoXlH5asQ5IVKTL9NLwUUkAXrWtWMFZRbKqy3OPXG0/056qUNfJAGmZ07Q2dHGkbmFF8o6O9qYnjlRk/+I49wTR/vvpJOCi0gC1KP5KM4JkZp8mT5qFhNJADUfSdIocxFJCDUfSZIouIgkSFqbj7TgZfoouIhIXWmGfjqpz0VE6ibtM/Sbec00ZS4iUjfZGfonOT0SLjtDP+nNY82ekSlzEZG6SesM/bRnZOVQcBGRuknrEGutmaZmMRGpszQOsU5rRrYSylxkxZq5k1Jqo7ezgwsGulMRWCC9GdlKKHORFWn2TkqRcqUxI1sJBRcpW24nZXZ0z8479rN1eH3T/eKIlCOtk17LoWYxKZs6KUWkXAouUjZ1UopIuRRcpGzqpBSRcqnPRVak2TspRaQ8Ci6yYs3cSSki5VGzmCRWJfNtNEdHJB7KXCSRKplvozk6IvFR5iINrVimUcmigKtdSFAZj8jKKHORhlUq0yg1ryZqmfZKrlmuHiJSmjIXaUhRmca69lZOzufPtzk5v8S69taSz1fJNcvVQ0RKU3CRhhS1GsCxU4u0FfzLbWuBY6cWSz7fsVOLdLRaXllHq0Ves1w9RKQ0BRdpSFGrAaxrb2Uh/xALS0RmIf09a7GW/OBiLbbs6gJalUCkMgou0pCiVgOoJAupdHWBeqxKoMEDkgbq0JeGVWo1gP6etRQkLiyF8kqer9J61IIGD0haKLhIQyu1GoC7Rz5e6fPV6rqV0JYGkiZqFpPEmZ45wdo1+Z+L1q5pS3wnuwYPSJrULLiY2c1m9qyZPZxT9nkz+4WZ7Tezb5lZd86xT5nZpJn90szekVO+LZRNmtk1OeVnm9n9ofybZtYeyjvC48lwfKhW71HqI62d7Gl9X9Kcapm5fA3YVlC2DzjX3c8HfgV8CsDMzgGuAF4XrvkLM2s1s1bgK8A7gXOA94dzAa4Drnf3YWAGuDKUXwnMhPLrw3mSItlO9o62Fs5sb6WjrfxO9ko7y+NYx0xbGkia1KzPxd1/WJg1uPs/5Dy8D7gsfL8duNXd54DHzWwSeGM4NunujwGY2a3AdjP7OXAR8IFwzi3AHwM3hOf641B+O/A/zMy83EZ5SQTP/un2wqPlVNpZHuc6ZtrSQNKinn0u/xb4Xvh+IzCVc2w6lJUq7wVm3X2hoDzvucLx58P5khLZju+5Bef4/CJzC16ztcXqsY5Zb2cHFwx0K7BIotUluJjZHwELwN/U4/Vz6nGVmY2b2fihQ4fqWRVZgUo6vivtLI/ztUTSJPbgYmYfBt4NfDCnqeoAMJBzWn8oK1V+GOg2s7aC8rznCsdfGs5/EXe/0d1H3X10w4YNq3xnEpdKOr4r7SyP87VE0iTW4GJm24CdwKXufjzn0F7gijDS62xgE/Aj4MfApjAyrJ1Mp//eEJTu4XSfzQ5gT85z7QjfXwbcrf6WdKmk4zvOGfrqmBcBq9X/u2b2DeCtwHrgIHAtmdFhHZzOJO5z938fzv8jMv0wC8Afuvv3Qvm7gD8HWoGb3f2zofzVwK3AWcBPgQ+5+5yZnQH8NfB64DngiuyAgCijo6M+Pj6++jcusTl8dG7FHd+VXBP3a4kkiZk94O6jLyrXh/oMBRcRkZUrFVw0Q19ERKpOwUVERKpOwUVERKpOwUVERKpOwUVWLGrNrCRvdJXkukdJ6/uSxqb9XGRFotbMSvJGV0mue5S0vi9pfMpcpGxRa2atdj2teqpH3ePIJpL8M5HkU+YiZcuumXUyZ5Ph3DWzSh1r9AmEUe+rFnWPK5uI+32J5FLmImWLWjOrHutpVevTf5x1jzOb0BpnUk8KLlK2qDWz4l5Pa8/EAbZedzcfuul+tl53N3snDix/UQlx1j3OFZN7Ozu4fLQ/r+zy0X5lLRILNYvJikRtZhXXRle5n/6zTT4779jP1uH1Fb9mXHWPO0u6bXw6r+y28Wk+dvHvK8BIzSlzkRWL2swqjo2uVvPpP6opLY66pzVLEimkzEUSp9JP/40yLDeNWZJIobIyFzN70b/+YmUilVhpx3wln/4bbVhuVJZUrYEK2ldG6qnczOVe4MIyykRWpNJsYqWf/pMyLLfa2VVcWZJIocjgYmYvBzYCa83s9YCFQy8Bzqxx3STlVtsxnx2lVo4kNBHVYqACrOw+iVTLcpnLO4APk9mj/os55UeAT9eoTtIkaplNFO4CmW0i2lmQFTTSf7pJya5EyhEZXNz9FuAWM3uvu98RU52kSdQqmyjVtNToTURJyK5EylXuUOTvmNkHzOzTZvaZ7FdNayaJtJLO6Fp0OC/XcR/HcONKqQNe0qTcDv09wPPAA4BWvZOiKumMrnY2kfSmpUbPrkTKVW5w6Xf3bTWtidRFYd/Eap6n0s7oanY4p6FpSR3wkgblNov9PzM7r6Y1kdhVc32uRpkNrqYlkcZQbubyZuDDZvY4mWYxA9zdz69ZzaSmqj3stZEyBjUtidRfucHlnTWthcSu2n0TjTbUV01LIvVVVnBx9yfN7M3AJnf/KzPbAHTWtmpSS7XINJQxiEhWuWuLXQtcDXwqFK0Bvl6rSknt1apvopGH+opIfMptFvtXwOuBnwC4+9Nm1lWzWkkslGmISK2UG1xOububmQOY2boa1klipL4JEamFZZvFzMzIzND/n0C3mX0E+EfgL2tdOam9ai3vnjV58Ai3j08xefBIzV9LRBrXsplLyFjeB3wc+B3wGuAz7r6v1pWT2qr28u6f+fZD7L7vqRcej20ZZNf282ryWiLS2MqdRPkTYNbd/4u7f1KBJfmqvXnW5MEjeYEFYPe9TzF58MiyrxWV7VRKWZJIfZXb5/Im4INm9iRwLFuoSZTJNT1zgsUlzytbXPKK57lMTM2WLN/U11VyTs2X/vFXJbOdSilLEqm/coPLO2paC4nd/MIi84v5wWV+0ZlfWKzo+UYGukuW96xrLzqnZn5hsWi2M7Z5iOG+ygYj1mrDLRFZmbKaxdz9yWJfta6c1M4Th4+vqHw5w31djG0ZzCsb2zLIcF9XyTk1pV6rVBZUjkZZ40yk2ZWbuUjKRGUaldq1/TzGNg8xMTXLyEB3XvZRbE5NqT6W1dShkdY4E2lm5XboS8pEZRqrfd7LRgeKPk/h7P2ede1YwTkWyiu13MoD6ugXiYcylyYWlWnEYXrmBJ0dbRyZW3ihrLOjbdUbe5VaeUAd/SLxUXBpcsN9XSsOKpVuMFZ4XS2bsApXHlBHv0i8FFxkRSr99F/quriW6U/69sciSaPgImWr9NN/1HVxLZ6pjn6ReNWsQ9/MbjazZ83s4Zyys8xsn5k9Gv7uCeVmZl82s0kz229mF+ZcsyOc/6iZ7cgpf4OZPRSu+XJYA63ka8jqVTrMd7nr4limX9sfi8SrlqPFvgZsKyi7BrjL3TcBd4XHkNnpclP4ugq4ATKBAriWzAoBbwSuzQkWNwAfyblu2zKvIavU37OWkwWTLE8uLC776b+/Zy1HczrtAY7OLcSeNVw6spF/vvoivv7v3sQ/X32ROvNFaqhmwcXdfwg8V1C8HbglfH8L8J6c8t2ecR+Z1ZdfQWZlgH3u/py7zwD7gG3h2Evc/T53d2B3wXMVew0pYqVDczO3u/TjYuuETTw1Q/5Z4KF8NSoZVqzNzETiEXefS5+7PxO+/w3QF77fCEzlnDcdyqLKp4uUR73Gi5jZVWQyJQYHB0udllp7Jg6w8/b9tLYYi0vO5y+L7pyfnjlBpvXxdKgwsxc6xUutivz3+58u+nx/v/9pLj7n5RXXXcOKRRpX3SZRhoyj8ANtrK/h7je6+6i7j27YsKGWVWk4h4/O8cm/fZC5hSWOn1pkbmGJT/ztg5FZQNR6ZFGrIr+s64yiz1eqvJy6V3NFZxGpvriDy8HQpEX4+9lQfgAYyDmvP5RFlfcXKY96jaZVrPnokaefLxooHnn6+ZLPE7UeWdSqyC9dWzxBLlW+HK0fJtL44g4ue4HsiK8dwJ6c8rEwamwz8Hxo2roTuMTMekJH/iXAneHY78xscxglNlbwXMVeoyntmTjA1uvu5kM33c/W6+5m70Q2BhcuvMIy5dHrkUUde1VvZ9FjpcqXo2HFIo2vlkORvwHcC7zGzKbN7ErgT4G3m9mjwNvCY4DvAo8Bk2S2T/4DAHd/DvgT4Mfha1coI5xzU7jm18D3Qnmp12g6Uc1Hr3xp8SapUuUQvR5Z1Dphr3158RUASpUvJzusuKOthTPbW+lo07BikUZTsw59d39/iUMXFznXgY+WeJ6bgZuLlI8D5xYpP1zsNZpR1Kx0gI5WYy6naayj1Th2Kno/l1LrkUWtEwbQYpC7N1mLsexrRfHsn54/wEBEGoNm6KfYcs1H1mKQE1ysxcpqWiq2HlnUaz1+6CgFm16y5FS8MVk2I5tbcCDzHFonTKSxaMn9FIualV7tGetRz/dwiUECpcqXow59kcanzCXlotbuqva6XqWer6Ottej5pcqXow59kcan4JISUcvgFy4/X+6xShR7vo09ZxY9t1R5Oa8R12rKIlIZBZcUaPTZ6q975Utoa4GFnGSjrSVTXqm4VlMWkcqozyXhypmtXu+tfXs7O/ji5SO0txodbS20txpfvHxk1QFB64SJNC5lLgm33CZYK10/rFYcMINWMxY1dFgk9ZS5JFxU53Yl64fVQu7Q4ePzi8wtuNYCE0k5BZeEixoCXM76YcWWyK82DR0WaT5qFkuB0p3b0euHlVoiv9o0dFik+ShzSYlindvZUVq5sqO0opbIh+oOAtAWwyLNR5lLivV2dvCBNw2y+97TQeQDbxqkt7ODe35RfCeCialZHnnmd1Uf2qyhwyLNRZlLih0+Osdt49N5ZbeNT3P46FzJJfKHes+s2UZcpYYO13uotIhUnzKXFIsaptzfs5bC9YQNOD6/GDm0udoafQKoiFRGmUuKRXWkZ5fIz5V5bLF1vmu7YpH0UnBJsahNtUoFnte98iU163wvbP7SEGWR9FKzWMqV2lSrt7ODy0f78zr7Lx/tp7ezoyad78Wav7YOr+f4fP6eLsfnFzVEWSQFlLmkRLFO8aiZ8VGd/atRbFJmqeavxw8dZbFgF7HFJWfm2KlV1UFE6k+ZSwqU6hRfbpvjUsf+7+RvK+pkLzUps1Q9fvjob4s+z8TU7It2uhSRZFHmknBRneJRHfqljq1rb62okz1qUmap13rLpvVFn6vUMGkRSQ4Fl4SL6hSP6tAvNWv+2KnFijrZJ6ZmS5aXeq3Rs3sZ2zKYd/7YlkFlLSIpoGaxhFtu3a5SHfpQfNb84aNznFzI72Q/ubB8J3upbCNbXmqQwK7t5zG2eYiJqVlGBroVWERSQplLwmVHfeXKjvoqZ6n7YrPm3fODUOHjYob7upbNQkrN0B/u6+Ky0QEFFpEUUeaScKVGfX3s4t9neuZE0dFYUbPtp2dOsHZNG0fmFl4oW7umrawZ+spCRCRLwSXhokaEzS8sFt3PZb6g2SvXapfHH+7rUlARETWLJV1UMHji8PGi1+SWF86PiRoEICJSLmUuCRc10365TvZS82OiBgGIiJRDmUvCRc20nz1efKb77PFTJefHTB48sqr97rV8voiAMpfEi+pzKTUD/oeP/pY1ba1Fr5uYmq14yX0tny8iWcpcEi6qz+Vlne1Fr3lZZ3vJ60YGuivq0Nfy+SKSS8El4aLmuTx7tHiz2LNHT5W8brivq6Il97V8vojkUrNYwkXNc3n1+jOLXvPq9WdGXlfJkvurHcIsIumizCXhojKGhaXi1ywsLZ9plJpNX0qp9cM0hFmkOSlzSbiojGFde2vRa0YGuulZ1171jbpqscmYiCSTMpeEi5r02LOuHSs434Cede3MHDtVk426VprxiEg6KbikwOlJj6cfQabpq2NN/o+4Y02m6StqiXwRkdVScEm4qJWP17W3cnI+v8ns5HxmQ7Ch3uKd/aXKRURWQsEl4aI65o+dWqSjNb9hrKPVMhuCtbVScIhWgzVtxftpRERWQh36CbfcEGBrMchZGdla7IVjLS3GYs6xlpxjIiKrUZfMxcz+s5k9YmYPm9k3zOwMMzvbzO43s0kz+6aZtYdzO8LjyXB8KOd5PhXKf2lm78gp3xbKJs3smjq8xdiUs5VxqRWOK9kUTESkHLEHFzPbCPwnYNTdzwVagSuA64Dr3X0YmAGuDJdcCcyE8uvDeZjZOeG61wHbgL8ws1YzawW+ArwTOAd4fzg3tUp16Ecdm545QVtr/o+/rVUz6kWkOurV59IGrDWzNuBM4BngIuD2cPwW4D3h++3hMeH4xWZmofxWd59z98eBSeCN4WvS3R9z91PAreHcVIrq0K+0s19EZLViDy7ufgD4AvAUmaDyPPAAMOvu2b11p4Hscrobgalw7UI4vze3vOCaUuUvYmZXmdm4mY0fOnRo9W+uDkplGtMzJ5bt7G8p6NBvMTh2qvQulSIi5apHs1gPmUzibOCVwDoyzVqxc/cb3X3U3Uc3bNhQjyqsWlQGEtXZP7+wSMEcSpacyC2QRUTKVY9msbcBj7v7IXefB/4O2Ap0h2YygH7gQPj+ADAAEI6/FDicW15wTanyVIoabhzVoV/OFsgiIpWqR3B5CthsZmeGvpOLgZ8B9wCXhXN2AHvC93vDY8Lxuz0zrGkvcEUYTXY2sAn4EfBjYFMYfdZOptN/bwzvqy76e9ZmhhvnyB1uXKpDf7ktkEVEVqMefS73k+mY/wnwUKjDjcDVwMfNbJJMn8pXwyVfBXpD+ceBa8LzPALcRiYwfR/4qLsvhn6Z/wjcCfwcuC2cm0q52Un2K5udRHXoD/d1MbZlMO+5xrYMMtzXVad3IiJpUpdJlO5+LXBtQfFjZEZ6FZ57Enhfief5LPDZIuXfBb67+pomw/gTzzGXs77++JPPcenIxsgtkHs7O9i1/TzGNg8xMTXLyEC3AouIVI1m6Cfc5MEj7L7vqbyy3fc+xdjmobI28Bru61JQEZGq09piCRe1urE28BKRelHmknDLdcxrAy8RqQdlLglXTse8NvASkbgpc0kBdcyLSKNRcEkJdcyLSCNRs5iIiFSdgouIiFSdgouIiFSdgouIiFSdgssqHT46x4NTsxw+OlfvqoiINAyNFluFPRMHuPqO/axpaWF+aYnPvfd8Lh0pui+ZiEhTUeZSoeyKwyfnlzgyt8DJ+aUXVhwWEWl2Ci4VitpCWESk2Sm4VKicFYdFRJqVgkuF6rHisAYPiEhSqEN/FeJccViDB0QkSRRcVqm3s6Pmqw3nDh7I7iq58479bB1er5WORaQhqVksATR4QESSRsElAfp71nJyYTGv7OTCogYPiEjDUnBJCHePfCwi0kgUXBJgeuYEZpZXZmZqFhORhqXgkgDzC4vML+ZnKvOLznxBU5mISKNQcEmAJw4fX1G5iEi9KbgkwMhA94rKRUTqTcElAYb7uhjbMphXNrZlkOG+rjrVSEQkmiZRJsSu7ecxtnmIialZRga6FVhEpKEpuCTIcF+XgoqIJIKaxUREpOoUXEREpOoUXEREpOoUXEREpOoUXEREpOpMCyBmmNkh4Ml612OV1gO/rXclGojux2m6F/l0P/Kt5n68yt03FBYquKSImY27+2i969EodD9O073Ip/uRrxb3Q81iIiJSdQouIiJSdQou6XJjvSvQYHQ/TtO9yKf7ka/q90N9LiIiUnXKXEREpOoUXEREpOoUXBLIzAbM7B4z+5mZPWJmHwvlZ5nZPjN7NPzdU++6xsHMzjCzH5nZg+F+/NdQfraZ3W9mk2b2TTNrr3dd42JmrWb2UzP7TnjczPfiCTN7yMwmzGw8lDXl7wqAmXWb2e1m9gsz+7mZbanF/VBwSaYF4BPufg6wGfiomZ0DXAPc5e6bgLvC42YwB1zk7hcAI8A2M9sMXAdc7+7DwAxwZf2qGLuPAT/PedzM9wLgX7j7SM5cjmb9XQH4EvB9d38tcAGZfydVvx8KLgnk7s+4+0/C90fI/OPYCGwHbgmn3QK8py4VjJlnHA0P14QvBy4Cbg/lTXM/zKwf+JfATeGx0aT3IkJT/q6Y2UuBtwBfBXD3U+4+Sw3uh4JLwpnZEPB64H6gz92fCYd+A/TVq15xC81AE8CzwD7g18Csuy+EU6bJBOBm8OfATmApPO6lee8FZD5o/IOZPWBmV4WyZv1dORs4BPxVaDa9yczWUYP7oeCSYGbWCdwB/KG7/y73mGfGmDfNOHN3X3T3EaAfeCPw2vrWqD7M7N3As+7+QL3r0kDe7O4XAu8k04T8ltyDTfa70gZcCNzg7q8HjlHQBFat+6HgklBmtoZMYPkbd/+7UHzQzF4Rjr+CzKf4phJS/HuALUC3mWW38u4HDtSrXjHaClxqZk8At5JpDvsSzXkvAHD3A+HvZ4Fvkfnw0ay/K9PAtLvfHx7fTibYVP1+KLgkUGhD/yrwc3f/Ys6hvcCO8P0OYE/cdasHM9tgZt3h+7XA28n0Q90DXBZOa4r74e6fcvd+dx8CrgDudvcP0oT3AsDM1plZV/Z74BLgYZr0d8XdfwNMmdlrQtHFwM+owf3QDP0EMrM3A/8EPMTpdvVPk+l3uQ0YJLN9wOXu/lxdKhkjMzufTCdkK5kPTLe5+y4zezWZT+9nAT8FPuTuc/WrabzM7K3AJ9393c16L8L7/lZ42Ab8L3f/rJn10oS/KwBmNkJmsEc78Bjwbwi/N1Txfii4iIhI1alZTEREqk7BRUREqk7BRUREqk7BRUREqk7BRUREqk7BRaTBhCU5zql3PURWQ0ORRUSk6pS5iNRRmEH+v8NeNA+b2b82sx+Y2aiZXRr2IJkws1+a2ePhmjeY2f8JCzHemV22Q6SRKLiI1Nc24Gl3v8DdzwW+nz3g7nvDHiQjwIPAF8Kacv8duMzd3wDcDHy2DvUWidS2/CkiUkMPAX9mZtcB33H3f8osHXeame0ETrj7V8zsXOBcYF84rxV4BpEGo+AiUkfu/iszuxB4F/DfzOyu3ONm9jbgfWQ2eAIw4BF33xJvTUVWRs1iInVkZq8Ejrv714HPk1n+PHvsVcBXgPe5+4lQ/Etgg5ltCeesMbPXxVxtkWUpcxGpr/OAz5vZEjAP/AfgC+HYh8nsIvnt0AT2tLu/y8wuA74ctqxtI7Pz5CMx11skkoYii4hI1alZTEREqk7BRUREqk7BRUREqk7BRUREqk7BRUREqk7BRUREqk7BRUREqu7/A9Fi4G6WKHsZAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "df_room.plot('size','rent', kind='scatter')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d223b0b1-870a-4908-ad2c-d8c931a18482",
   "metadata": {},
   "source": [
    "Pandasデータフレームのlocメソッドで列を取り出します。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "7ae664e8-ad6b-426a-b058-a7d4e08db7ec",
   "metadata": {},
   "outputs": [],
   "source": [
    "x = df_room.loc[:,'size']\n",
    "y = df_room.loc[:,'rent']"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "36312ea5-919c-4753-9b8a-bfc3ffcb2bab",
   "metadata": {},
   "source": [
    "切片を計算させるために定数項を設定します。  \n",
    "**注意**：もし切片を必要とする線形回帰のモデル式ならば必ず必要な操作で、これを入れないと正しく回帰式が作成されません。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "a34b421f-23bf-477c-a3e0-05f2f83166fe",
   "metadata": {},
   "outputs": [],
   "source": [
    "X1 = sm.add_constant(x)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cf7c1b99-8445-4921-99cd-40793667b53b",
   "metadata": {},
   "source": [
    "上の操作によって、次に示すように、すべてのデータに先頭項目として定数１が加わりました。  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "id": "6fbec026-bd1d-45d7-9b3d-22d87e076e42",
   "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>const</th>\n",
       "      <th>size</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1.0</td>\n",
       "      <td>19</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1.0</td>\n",
       "      <td>19</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1.0</td>\n",
       "      <td>19</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1.0</td>\n",
       "      <td>19</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1.0</td>\n",
       "      <td>15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>135</th>\n",
       "      <td>1.0</td>\n",
       "      <td>40</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>136</th>\n",
       "      <td>1.0</td>\n",
       "      <td>54</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>137</th>\n",
       "      <td>1.0</td>\n",
       "      <td>42</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>138</th>\n",
       "      <td>1.0</td>\n",
       "      <td>42</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>139</th>\n",
       "      <td>1.0</td>\n",
       "      <td>41</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>140 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     const  size\n",
       "0      1.0    19\n",
       "1      1.0    19\n",
       "2      1.0    19\n",
       "3      1.0    19\n",
       "4      1.0    15\n",
       "..     ...   ...\n",
       "135    1.0    40\n",
       "136    1.0    54\n",
       "137    1.0    42\n",
       "138    1.0    42\n",
       "139    1.0    41\n",
       "\n",
       "[140 rows x 2 columns]"
      ]
     },
     "execution_count": 71,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X1"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8f65459d-65dc-488f-9287-bb3fe814c989",
   "metadata": {},
   "source": [
    "モデルを設定します。OLSとは、最小二乗法のことです。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "47d2c7cd-9f74-4581-b388-90400611cda7",
   "metadata": {},
   "outputs": [],
   "source": [
    "model_1 = sm.OLS(y,X1)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "58389601-dac1-4da6-9d9b-30195532f704",
   "metadata": {},
   "source": [
    "回帰分析を実行します。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "56e4dd79-13d3-4377-a6f5-46f10d77bedf",
   "metadata": {},
   "outputs": [],
   "source": [
    "result_1 = model_1.fit()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6cc3b98e-cb69-4844-945b-da05e2f43399",
   "metadata": {},
   "source": [
    "結果の詳細を表示します。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "id": "488016f8-1ff6-49c5-813c-81b70bfbacdb",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                            OLS Regression Results                            \n",
      "==============================================================================\n",
      "Dep. Variable:                   rent   R-squared:                       0.708\n",
      "Model:                            OLS   Adj. R-squared:                  0.705\n",
      "Method:                 Least Squares   F-statistic:                     334.0\n",
      "Date:                Sun, 22 May 2022   Prob (F-statistic):           1.14e-38\n",
      "Time:                        15:42:46   Log-Likelihood:                -1493.4\n",
      "No. Observations:                 140   AIC:                             2991.\n",
      "Df Residuals:                     138   BIC:                             2997.\n",
      "Df Model:                           1                                         \n",
      "Covariance Type:            nonrobust                                         \n",
      "==============================================================================\n",
      "                 coef    std err          t      P>|t|      [0.025      0.975]\n",
      "------------------------------------------------------------------------------\n",
      "const       4.579e+04   3179.976     14.400      0.000    3.95e+04    5.21e+04\n",
      "size        2075.1455    113.553     18.275      0.000    1850.617    2299.674\n",
      "==============================================================================\n",
      "Omnibus:                        6.374   Durbin-Watson:                   1.308\n",
      "Prob(Omnibus):                  0.041   Jarque-Bera (JB):                6.691\n",
      "Skew:                           0.352   Prob(JB):                       0.0352\n",
      "Kurtosis:                       3.808   Cond. No.                         101.\n",
      "==============================================================================\n",
      "\n",
      "Notes:\n",
      "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n"
     ]
    }
   ],
   "source": [
    "print(result_1.summary())"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "872bc464-db7e-4285-a503-7549edf17469",
   "metadata": {},
   "source": [
    "表の解説の前に、回帰直線を表示しましょう。パラメータ$\\hat{\\alpha}$, $\\hat{\\beta}$（それぞれ、y切片と傾き）は結果から次のように取り出されます。解説より先に、上の表でいうと、それぞれ、const行のcoef, size行のcoefにあたります。これらの値は、回帰直線をプロットするために使います。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "id": "e32abd8b-f235-43f4-8aff-8e4c9df98adb",
   "metadata": {},
   "outputs": [],
   "source": [
    "a_1 = result_1.params[0] #alpha\n",
    "b_1 = result_1.params[1] #beta"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9a98decf-189f-4a6c-aac9-4382753fea0f",
   "metadata": {},
   "source": [
    "ではデータをプロットし、回帰直線を表示します。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "id": "3e0421e5-e646-47f5-95e0-1a00918632c6",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(x, y,\"o\")\n",
    "plt.plot(x,a_1 + b_1*x)\n",
    "plt.xlabel('size')\n",
    "plt.ylabel('rent')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "99ea87b9-bb16-49eb-9d56-b996ae82a880",
   "metadata": {},
   "source": [
    "上の表の内容を説明します。まず再掲します。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "id": "a8188846-c35a-4afe-89a0-0c6c7cca3951",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                            OLS Regression Results                            \n",
      "==============================================================================\n",
      "Dep. Variable:                   rent   R-squared:                       0.708\n",
      "Model:                            OLS   Adj. R-squared:                  0.705\n",
      "Method:                 Least Squares   F-statistic:                     334.0\n",
      "Date:                Sun, 22 May 2022   Prob (F-statistic):           1.14e-38\n",
      "Time:                        16:06:50   Log-Likelihood:                -1493.4\n",
      "No. Observations:                 140   AIC:                             2991.\n",
      "Df Residuals:                     138   BIC:                             2997.\n",
      "Df Model:                           1                                         \n",
      "Covariance Type:            nonrobust                                         \n",
      "==============================================================================\n",
      "                 coef    std err          t      P>|t|      [0.025      0.975]\n",
      "------------------------------------------------------------------------------\n",
      "const       4.579e+04   3179.976     14.400      0.000    3.95e+04    5.21e+04\n",
      "size        2075.1455    113.553     18.275      0.000    1850.617    2299.674\n",
      "==============================================================================\n",
      "Omnibus:                        6.374   Durbin-Watson:                   1.308\n",
      "Prob(Omnibus):                  0.041   Jarque-Bera (JB):                6.691\n",
      "Skew:                           0.352   Prob(JB):                       0.0352\n",
      "Kurtosis:                       3.808   Cond. No.                         101.\n",
      "==============================================================================\n",
      "\n",
      "Notes:\n",
      "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n"
     ]
    }
   ],
   "source": [
    "print(result_1.summary())"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f7e15099-456f-4e19-a5a8-c2f2e3cefccd",
   "metadata": {},
   "source": [
    "項目をかいつまんで説明します。  \n",
    "- Dep. Variable: rent       ：従属変数は'rent'です。  \n",
    "- Model: OLS                ：モデル化のために最小二乗法（Ordinary Least Square）を使いました。  \n",
    "- No. Observations: 140     ：データの数です。  \n",
    "- Df Residuals: 138         ：データの自由度です。  \n",
    "- Df Model: 1               ：モデルの自由度です。$\\hat{\\alpha}$と$\\hat{\\beta}$の２つの変数なので、2-1 = 1。  \n",
    "- R-squared: 0.708          ：上で示した決定係数$R^2$が0.708です。  \n",
    "- Adj. R-squared: 0.705     ：重回帰分析のときに効いてくる指標です。説明変数を増やすとペナルティとして$R^2$を減らされます。   \n",
    "- F-statistics　　　　　　  ：データとモデルの自由度を考慮して、$S_R$と$S_e$を比較する指標です。大きいほど帰無仮説（回帰係数が0）のもとで稀な事象となります。\n",
    "- Prob(F-statistics)　　　　：帰無仮説のもとで上記F-statistics以上の値を得る確率です。\n",
    "- AIC: ：モデルへの当てはまりの良さを表します。小さいほど良いとされています。  \n",
    "- const とsizeのcoef: ：上述した、回帰係数です。  \n",
    "- std err: ：回帰係数の標準誤差です。  \n",
    "- t: ：帰無仮説のもとでのt統計値です。  \n",
    "- P>|t| ：P-値です。統計値がこの値よりも極端な値となる確率です。。  \n",
    "-  \\[0.025      0.975\\]: ：回帰係数$\\hat{\\beta}$の95%信頼区間です。"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "71387a2f-be7d-4db2-bc3f-17689eddfd53",
   "metadata": {},
   "source": [
    "#### 例題"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b00bf22e-1f3b-4cf7-a8cd-b2752f235d48",
   "metadata": {},
   "source": [
    "さきほどの、身長と体重との関係について、次のステップで回帰分析を行いましょう。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "083c7ce9-4c2f-40f6-b6c6-f4b0c1133895",
   "metadata": {},
   "outputs": [],
   "source": [
    "import statsmodels.formula.api as smf\n",
    "import statsmodels.api as sm"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "689854ba-78d7-474b-9cea-b239f874f117",
   "metadata": {},
   "source": [
    "データフレームdf_hwからxとyに相当するもの（それぞれ身長hと体重w）をデータフレームから取り出しましょう。.loc[;,列名]メソッドで取り出します。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "a4ae3f0b-135e-47c2-9105-e5e7af759756",
   "metadata": {},
   "outputs": [],
   "source": [
    "h = df_hw.loc[:,'Height']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "6c44460b-d417-4dab-9d3e-f0b50f2cefbb",
   "metadata": {},
   "outputs": [],
   "source": [
    "w = df_hw.loc[:,'Weight']"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "aa4f9d6f-145a-46e2-ace1-2bd199b1bf6a",
   "metadata": {},
   "source": [
    "身長hに定数項を設定しましょう。sm.add_constant()関数を用います。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "d0ddecac-e917-4c5b-8559-914334347e27",
   "metadata": {},
   "outputs": [],
   "source": [
    "h1 = sm.add_constant(h)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c7c1ccf8-515d-4d1a-bea4-f990686a4363",
   "metadata": {},
   "source": [
    "モデルを設定しましょう。sm.OLS(y,x)関数を使います。xには必要に応じて定数項が付加されている必要があります。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "2e8cc135-bf5a-4923-8dbd-8cb3238e61c9",
   "metadata": {},
   "outputs": [],
   "source": [
    "model_hw = sm.OLS(w,h1)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "33d36f7f-e265-4d69-a186-902f12e2d812",
   "metadata": {},
   "source": [
    "回帰分析を実行して、結果の詳細を表示しましょう。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "40777d78-7f21-44e4-8dd5-a12c22b63303",
   "metadata": {},
   "outputs": [],
   "source": [
    "result_hw = model_hw.fit()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "2d2159e8-56b0-49d3-9213-898e1352138e",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                            OLS Regression Results                            \n",
      "==============================================================================\n",
      "Dep. Variable:                 Weight   R-squared:                       0.465\n",
      "Model:                            OLS   Adj. R-squared:                  0.398\n",
      "Method:                 Least Squares   F-statistic:                     6.956\n",
      "Date:                Fri, 01 Jul 2022   Prob (F-statistic):             0.0298\n",
      "Time:                        01:25:52   Log-Likelihood:                -29.175\n",
      "No. Observations:                  10   AIC:                             62.35\n",
      "Df Residuals:                       8   BIC:                             62.95\n",
      "Df Model:                           1                                         \n",
      "Covariance Type:            nonrobust                                         \n",
      "==============================================================================\n",
      "                 coef    std err          t      P>|t|      [0.025      0.975]\n",
      "------------------------------------------------------------------------------\n",
      "const        -61.8937     43.743     -1.415      0.195    -162.766      38.979\n",
      "Height         0.7219      0.274      2.637      0.030       0.091       1.353\n",
      "==============================================================================\n",
      "Omnibus:                        0.969   Durbin-Watson:                   2.006\n",
      "Prob(Omnibus):                  0.616   Jarque-Bera (JB):                0.622\n",
      "Skew:                          -0.005   Prob(JB):                        0.733\n",
      "Kurtosis:                       1.778   Cond. No.                     4.42e+03\n",
      "==============================================================================\n",
      "\n",
      "Notes:\n",
      "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
      "[2] The condition number is large, 4.42e+03. This might indicate that there are\n",
      "strong multicollinearity or other numerical problems.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "E:\\PythonUSB\\PythonUSB\\Python\\python-3.9.9\\lib\\site-packages\\scipy\\stats\\stats.py:1541: UserWarning: kurtosistest only valid for n>=20 ... continuing anyway, n=10\n",
      "  warnings.warn(\"kurtosistest only valid for n>=20 ... continuing \"\n"
     ]
    }
   ],
   "source": [
    "print(result_hw.summary())"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "804eec29-bcb9-46eb-b21e-995ee2eeec06",
   "metadata": {},
   "source": [
    "$R^2$が0.465と、それほどは高くはないことがわかります。身長以外の要因もかなり関係するであろうことがわかります。"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ff7e3dbe-1e3d-44d8-9c1d-1f10bc1018bf",
   "metadata": {},
   "source": [
    "しかし、$\\beta$つまり身長$h$の係数は正で、$P-$値は0.030つまり5%水準で有意で、95%信頼区間は正の範囲にあります。要するに、身長は体重に関係していると言ってよいことがわかります。"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0fa30a43-215f-458b-bdd5-f7731cefe77b",
   "metadata": {},
   "source": [
    "回帰モデルのパラメータを抽出しましょう。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "06f52432-e2b2-41bd-9776-ca61fbb84249",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "-61.893744387907816"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "a_2 = result_hw.params[0]\n",
    "a_2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "78f7bcbd-377c-4626-8a65-4623cdc3cc6e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.7219395390601615"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "b_2 = result_hw.params[1]\n",
    "b_2"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7048d06f-9dbb-4cb5-9d0d-4611b6c35f08",
   "metadata": {},
   "source": [
    "データをプロットし、回帰直線を図示しましょう。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "bdd21bc0-bc61-40a5-ab1f-aec3f9ba165a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(h, w,\"o\")\n",
    "plt.plot(h,a_2 + b_2*h)\n",
    "plt.xlabel('Height')\n",
    "plt.ylabel('Weight')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "6ee45e45-588f-4aae-9c12-d7af31cdd1e6",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "id": "b06e21d6-90b3-46c3-8e71-115ece5be246",
   "metadata": {
    "tags": []
   },
   "source": [
    "### 補足：線形モデルの妥当性"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1d651daa-8fec-4a4d-a062-ba54a74d0196",
   "metadata": {},
   "source": [
    "上に示したサマリーの、下のほうの欄は、線形モデルを使うことの適切性についての指標です。"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b8479ecd-2bb2-4009-bff3-b1f723b08706",
   "metadata": {},
   "source": [
    "Omnibus, Prob(Omnibus): ：Omnibusの値が小さく、Prob(Omnibus)の値が1に近いほど、残差の分布が正規分布に近くなります。  \n",
    "とはいえ、具体的にどれだけであるべきかについての決まりはありません。  \n",
    "そこで、残差をヒストグラムにして調べてみます。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 92,
   "id": "a6cfac0f-dc4b-4ea8-a725-e4bf7714f9f6",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(array([ 2.,  0., 12., 32., 36., 30., 13., 13.,  0.,  2.]),\n",
       " array([-31300.17283685, -24774.73636911, -18249.29990138, -11723.86343365,\n",
       "         -5198.42696591,   1327.00950182,   7852.44596955,  14377.88243729,\n",
       "         20903.31890502,  27428.75537275,  33954.19184049]),\n",
       " <BarContainer object of 10 artists>)"
      ]
     },
     "execution_count": 92,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.hist(result_1.resid)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6691a607-8fa7-49a8-8cfc-157f09894333",
   "metadata": {},
   "source": [
    "残差のskewness（歪度：正に大きいほど右すそが大きく、負に大きいほど左すそが大きい）とkurtosis（尖度：大きいほど分布がとがる）は、それぞれ、SkewとKurtosisに表されています。"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "db10883c-de81-4d97-8d2e-71a1c89e7358",
   "metadata": {},
   "source": [
    "次に、sizeと残差の関係を散布図で見てみましょう。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 90,
   "id": "dddcf92b-b7f2-453d-8a83-95140c78230b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.collections.PathCollection at 0x227d87c0940>"
      ]
     },
     "execution_count": 90,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.scatter(x,result_1.resid)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "15bd924e-e5f1-4ec1-a883-d24ac097c458",
   "metadata": {},
   "source": [
    "sizeが大きくなるほど残差の絶対値が大きくなる傾向が見て取れます。"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6981b430-9281-4eed-9763-377784c43f22",
   "metadata": {
    "tags": []
   },
   "source": [
    "### さらに補足：lmdiagを使った線形モデルの妥当性の検討"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2fdf889b-931e-47cb-9a8b-476130101f69",
   "metadata": {},
   "source": [
    "lmdiagというツールで、線形モデルの妥当性を検討することができます。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 94,
   "id": "d1c59f10-c18a-4867-9a77-68a29416956b",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Note: you may need to restart the kernel to use updated packages.Collecting lmdiag\n",
      "  Downloading lmdiag-0.3.7-py3-none-any.whl (7.9 kB)\n",
      "Collecting linearmodels\n",
      "\n",
      "  Downloading linearmodels-4.26-cp39-cp39-win_amd64.whl (1.6 MB)\n",
      "     ---------------------------------------- 1.6/1.6 MB 7.8 MB/s eta 0:00:00\n",
      "Requirement already satisfied: pandas in e:\\pythonusb\\pythonusb\\python\\python-3.9.9\\lib\\site-packages (from lmdiag) (1.4.2)\n",
      "Requirement already satisfied: statsmodels in e:\\pythonusb\\pythonusb\\python\\python-3.9.9\\lib\\site-packages (from lmdiag) (0.13.2)\n",
      "Requirement already satisfied: matplotlib in e:\\pythonusb\\pythonusb\\python\\python-3.9.9\\lib\\site-packages (from lmdiag) (3.5.0)\n",
      "Requirement already satisfied: numpy in e:\\pythonusb\\pythonusb\\python\\python-3.9.9\\lib\\site-packages (from lmdiag) (1.21.4)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING: You are using pip version 22.0.4; however, version 22.1 is available.\n",
      "You should consider upgrading via the 'E:\\PythonUSB\\PythonUSB\\Python\\python-3.9.9\\pythonw.exe -m pip install --upgrade pip' command.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Requirement already satisfied: scipy in e:\\pythonusb\\pythonusb\\python\\python-3.9.9\\lib\\site-packages (from lmdiag) (1.7.3)\n",
      "Collecting Cython>=0.29.21\n",
      "  Downloading Cython-0.29.30-py2.py3-none-any.whl (985 kB)\n",
      "     -------------------------------------- 985.1/985.1 KB 8.9 MB/s eta 0:00:00\n",
      "Collecting mypy-extensions>=0.4\n",
      "  Downloading mypy_extensions-0.4.3-py2.py3-none-any.whl (4.5 kB)\n",
      "Collecting formulaic~=0.3.2\n",
      "  Downloading formulaic-0.3.4-py3-none-any.whl (68 kB)\n",
      "     ---------------------------------------- 68.2/68.2 KB 3.6 MB/s eta 0:00:00\n",
      "Collecting pyhdfe>=0.1\n",
      "  Downloading pyhdfe-0.1.0-py3-none-any.whl (18 kB)\n",
      "Collecting property-cached>=1.6.3\n",
      "  Downloading property_cached-1.6.4-py2.py3-none-any.whl (7.8 kB)\n",
      "Requirement already satisfied: patsy in e:\\pythonusb\\pythonusb\\python\\python-3.9.9\\lib\\site-packages (from linearmodels->lmdiag) (0.5.2)\n",
      "Collecting setuptools-scm<7.0.0,>=6.4.2\n",
      "  Downloading setuptools_scm-6.4.2-py3-none-any.whl (37 kB)\n",
      "Requirement already satisfied: python-dateutil>=2.8.1 in e:\\pythonusb\\pythonusb\\python\\python-3.9.9\\lib\\site-packages (from pandas->lmdiag) (2.8.2)\n",
      "Requirement already satisfied: pytz>=2020.1 in e:\\pythonusb\\pythonusb\\python\\python-3.9.9\\lib\\site-packages (from pandas->lmdiag) (2021.3)\n",
      "Requirement already satisfied: packaging>=21.3 in e:\\pythonusb\\pythonusb\\python\\python-3.9.9\\lib\\site-packages (from statsmodels->lmdiag) (21.3)\n",
      "Requirement already satisfied: cycler>=0.10 in e:\\pythonusb\\pythonusb\\python\\python-3.9.9\\lib\\site-packages (from matplotlib->lmdiag) (0.11.0)\n",
      "Requirement already satisfied: kiwisolver>=1.0.1 in e:\\pythonusb\\pythonusb\\python\\python-3.9.9\\lib\\site-packages (from matplotlib->lmdiag) (1.3.2)\n",
      "Requirement already satisfied: pillow>=6.2.0 in e:\\pythonusb\\pythonusb\\python\\python-3.9.9\\lib\\site-packages (from matplotlib->lmdiag) (8.4.0)\n",
      "Requirement already satisfied: fonttools>=4.22.0 in e:\\pythonusb\\pythonusb\\python\\python-3.9.9\\lib\\site-packages (from matplotlib->lmdiag) (4.28.2)\n",
      "Requirement already satisfied: pyparsing>=2.2.1 in e:\\pythonusb\\pythonusb\\python\\python-3.9.9\\lib\\site-packages (from matplotlib->lmdiag) (3.0.6)\n",
      "Collecting astor>=0.8\n",
      "  Downloading astor-0.8.1-py2.py3-none-any.whl (27 kB)\n",
      "Collecting wrapt>=1.0\n",
      "  Downloading wrapt-1.14.1-cp39-cp39-win_amd64.whl (35 kB)\n",
      "Collecting interface-meta<2.0.0,>=1.2.0\n",
      "  Downloading interface_meta-1.3.0-py3-none-any.whl (14 kB)\n",
      "Requirement already satisfied: six in e:\\pythonusb\\pythonusb\\python\\python-3.9.9\\lib\\site-packages (from patsy->linearmodels->lmdiag) (1.16.0)\n",
      "Requirement already satisfied: tomli>=1.0.0 in e:\\pythonusb\\pythonusb\\python\\python-3.9.9\\lib\\site-packages (from setuptools-scm<7.0.0,>=6.4.2->linearmodels->lmdiag) (1.2.2)\n",
      "Requirement already satisfied: setuptools in e:\\pythonusb\\pythonusb\\python\\python-3.9.9\\lib\\site-packages (from setuptools-scm<7.0.0,>=6.4.2->linearmodels->lmdiag) (59.2.0)\n",
      "Installing collected packages: mypy-extensions, wrapt, property-cached, interface-meta, Cython, astor, setuptools-scm, pyhdfe, formulaic, linearmodels, lmdiag\n",
      "  Attempting uninstall: setuptools-scm\n",
      "    Found existing installation: setuptools-scm 6.3.2\n",
      "    Uninstalling setuptools-scm-6.3.2:\n",
      "      Successfully uninstalled setuptools-scm-6.3.2\n",
      "Successfully installed Cython-0.29.30 astor-0.8.1 formulaic-0.3.4 interface-meta-1.3.0 linearmodels-4.26 lmdiag-0.3.7 mypy-extensions-0.4.3 property-cached-1.6.4 pyhdfe-0.1.0 setuptools-scm-6.4.2 wrapt-1.14.1\n"
     ]
    }
   ],
   "source": [
    "pip install lmdiag"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "b0b40f1d-f64d-4b09-b143-3ff22ec2710b",
   "metadata": {},
   "outputs": [],
   "source": [
    "import lmdiag"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "62a9df01-5d5e-4a19-a50d-525481925892",
   "metadata": {},
   "source": [
    "lmdiag.resid_fit()関数で、理想的な残差からどの程度乖離しているか視覚的にチェックすることができます。  \n",
    "出力されるグラフは、横軸が予測値、縦軸が残差です。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "27b0f06f-004e-4456-a059-80ac70b2cc4a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<module 'matplotlib.pyplot' from 'E:\\\\PythonUSB\\\\PythonUSB\\\\Python\\\\python-3.9.9\\\\lib\\\\site-packages\\\\matplotlib\\\\pyplot.py'>"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "lmdiag.resid_fit(result_1)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "61d3635a-8a7e-47b6-a6aa-49d6dbfc76d5",
   "metadata": {},
   "source": [
    "また、QQプロットというものも行われます。Qとは、quantile（分位数）のことです。これは、横軸に正規分布の理論値を、縦軸に、標準化残差（standardized residuals）をプロットしたものです。標準化残差とは平均0，分散1に変換した残差のことです。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "8397cea5-b96d-403a-a976-1a0ff463011f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<module 'matplotlib.pyplot' from 'E:\\\\PythonUSB\\\\PythonUSB\\\\Python\\\\python-3.9.9\\\\lib\\\\site-packages\\\\matplotlib\\\\pyplot.py'>"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "lmdiag.q_q(result_1)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3d16e690-d670-41b5-b4d9-b29034e58022",
   "metadata": {},
   "source": [
    "この結果が、上の赤い点線で示した45°の線上に乗っていれば残差は正規分布に従って分布しています。しかし、多くの点がそこから外れていると、残差が正規分布するといえなくなり、線形モデルの適用を見直す必要があります。"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b9f53845-675b-4dfb-9528-10baafdc9243",
   "metadata": {},
   "source": [
    "StatsmodelsにもQQプロットをする機能があります。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "48fe80f0-bb73-4003-95ef-cfa7998b91c5",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sm.qqplot(result_1.resid,line=\"s\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e4147479-e87e-4d2e-ae47-34b9a7b12b67",
   "metadata": {},
   "source": [
    "## 重回帰分析"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0db7b1b8-6f0b-4fba-81dc-609b4987e76c",
   "metadata": {},
   "source": [
    "これまで、説明変数が１個の単回帰分析を扱ってきました。説明変数が２個以上の場合、重回帰分析といいます。"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e4bc6173-f92d-4925-8ace-e6f9f8c3b4ad",
   "metadata": {},
   "source": [
    "説明変数が$p$個のとき、線形重回帰モデルは、$$y_i = \\alpha + \\beta_1 x_i + \\beta_2 x_i + ... + \\beta_p x_i + e_i$$ ただし$i = 1, 2, ... n$ ($n$はデータの数)となります。"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cbaf18d5-b38b-44ee-9205-a81d4ccc3037",
   "metadata": {},
   "source": [
    "データから、当てはめ値$\\hat{\\alpha}, \\hat{\\beta}_1, ,\\hat{\\beta}_2, ... \\hat{\\beta}_p$を求めることができます。"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7b7b72dc-4aba-44c4-a668-453f7ee98f98",
   "metadata": {},
   "source": [
    "### Statsmodelsを用いた線形重回帰分析"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1b67b85b-7c01-494d-9b20-2544d6255a65",
   "metadata": {},
   "source": [
    "#### 量的変数を加えた場合"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5d7c7a9c-c5d6-4945-97e3-536f0d288c73",
   "metadata": {},
   "source": [
    "家賃の線形モデルに築年数を加えてみます。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "a524c9af-1dbb-4d92-b3f5-1de0300653f3",
   "metadata": {},
   "outputs": [],
   "source": [
    "x_2 = df_room.loc[:,['size','age']]\n",
    "y = df_room.loc[:,'rent']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "d7b701a5-fdd6-4c24-8824-0fbcc63d8bf4",
   "metadata": {},
   "outputs": [],
   "source": [
    "X2 = sm.add_constant(x_2) #定数項の設定（先頭に値１のカラムを加える）"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "63fac280-0014-4c32-8fe7-e2b3fb364de0",
   "metadata": {},
   "outputs": [],
   "source": [
    "model_2 = sm.OLS(y,X2) # モデルの設定"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "2f4b1b73-4f36-426d-a00c-4a6808082649",
   "metadata": {},
   "outputs": [],
   "source": [
    "result_2 = model_2.fit() # 重回帰分析の実行"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "6111aef5-f482-43fe-a963-51663ba9c152",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                            OLS Regression Results                            \n",
      "==============================================================================\n",
      "Dep. Variable:                   rent   R-squared:                       0.756\n",
      "Model:                            OLS   Adj. R-squared:                  0.752\n",
      "Method:                 Least Squares   F-statistic:                     212.3\n",
      "Date:                Mon, 23 May 2022   Prob (F-statistic):           1.07e-42\n",
      "Time:                        09:41:44   Log-Likelihood:                -1480.7\n",
      "No. Observations:                 140   AIC:                             2967.\n",
      "Df Residuals:                     137   BIC:                             2976.\n",
      "Df Model:                           2                                         \n",
      "Covariance Type:            nonrobust                                         \n",
      "==============================================================================\n",
      "                 coef    std err          t      P>|t|      [0.025      0.975]\n",
      "------------------------------------------------------------------------------\n",
      "const       4.552e+04   2915.615     15.613      0.000    3.98e+04    5.13e+04\n",
      "size        2402.2011    121.520     19.768      0.000    2161.903    2642.499\n",
      "age         -892.4534    171.086     -5.216      0.000   -1230.765    -554.142\n",
      "==============================================================================\n",
      "Omnibus:                        6.066   Durbin-Watson:                   1.469\n",
      "Prob(Omnibus):                  0.048   Jarque-Bera (JB):                8.993\n",
      "Skew:                           0.137   Prob(JB):                       0.0111\n",
      "Kurtosis:                       4.211   Cond. No.                         107.\n",
      "==============================================================================\n",
      "\n",
      "Notes:\n",
      "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n"
     ]
    }
   ],
   "source": [
    "print(result_2.summary())"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "92dbfa77-3c96-4289-a5b2-bdbc9a71bfbb",
   "metadata": {},
   "source": [
    "この分析から、予測式は、家賃\\[円\\] = 45523.0 + 2402.2 x 広さ + (-892.5) x 築年数  \n",
    "となります。どちらの説明変数もt値の絶対値が十分大きく、従ってp値が0に近く、十分に有意です。  \n",
    "広さのみを説明変数とした場合とくらべて、Adjusted R-squaredの値が大きいのがわかります。"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7e4761b8-11d2-43fa-ba0e-890d17f1fcef",
   "metadata": {
    "tags": []
   },
   "source": [
    "#### 質的変数（ダミー変数）を加えた場合"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "de1c983e-f40f-4946-9692-9ef61e1981a6",
   "metadata": {},
   "source": [
    "家賃データで、近さproximityは、Aのときに「近い」、Bのときに「遠い」を表します。そこで、Aに0を、Bに1を割り当てることで質的変数をあたかも量的変数のように扱うことができます。このような変数のことを**ダミー変数**といいます。"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0ec56841-d88a-4fd6-81bc-564123f61005",
   "metadata": {},
   "source": [
    "df_roomデータフレームから説明変数としてsizeとproximityを選び、proximityをダミー変数化します。  \n",
    "pandas.get_dummies()関数を用います。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "768b3f11-3dcf-4541-8993-f35f06275d89",
   "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>size</th>\n",
       "      <th>proximity_A</th>\n",
       "      <th>proximity_B</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>19</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>19</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>19</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>19</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>15</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>135</th>\n",
       "      <td>40</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>136</th>\n",
       "      <td>54</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>137</th>\n",
       "      <td>42</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>138</th>\n",
       "      <td>42</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>139</th>\n",
       "      <td>41</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>140 rows × 3 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     size  proximity_A  proximity_B\n",
       "0      19            0            1\n",
       "1      19            0            1\n",
       "2      19            0            1\n",
       "3      19            0            1\n",
       "4      15            0            1\n",
       "..    ...          ...          ...\n",
       "135    40            0            1\n",
       "136    54            0            1\n",
       "137    42            1            0\n",
       "138    42            1            0\n",
       "139    41            0            1\n",
       "\n",
       "[140 rows x 3 columns]"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.get_dummies(df_room[['size','proximity']])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "680d51b8-a78c-47d1-ae96-535eced7e13e",
   "metadata": {},
   "source": [
    "上がその結果得られるデータフレームです。問題は、近さの指標としてproximity_Aかproximity_Bのどちらを使うかです。  \n",
    "遠いと家賃が下がることを想定しているので、遠いとフラグが立つほう、つまり、proximity_Bを選びます。\n",
    "説明変数としてsizeとproximity_Bを採用し、また、先頭に定数カラムをつくります。  \n",
    "次の命令では、loc[]の使い方にも注意してください。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "4714e001-fd33-405c-943d-556c8130bb23",
   "metadata": {},
   "outputs": [],
   "source": [
    "x_3 = pd.get_dummies(df_room[['size','proximity']]).loc[:,['size','proximity_B']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "9a4a4ad6-88f3-4f3d-928c-12cc8172a3d2",
   "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>size</th>\n",
       "      <th>proximity_B</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>19</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>19</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>19</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>19</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>15</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>135</th>\n",
       "      <td>40</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>136</th>\n",
       "      <td>54</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>137</th>\n",
       "      <td>42</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>138</th>\n",
       "      <td>42</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>139</th>\n",
       "      <td>41</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>140 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     size  proximity_B\n",
       "0      19            1\n",
       "1      19            1\n",
       "2      19            1\n",
       "3      19            1\n",
       "4      15            1\n",
       "..    ...          ...\n",
       "135    40            1\n",
       "136    54            1\n",
       "137    42            0\n",
       "138    42            0\n",
       "139    41            1\n",
       "\n",
       "[140 rows x 2 columns]"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x_3"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "c3f48f25-71cf-4f3c-979a-c7c20d3080a3",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "X3 = sm.add_constant(x_3)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "167e2798-88be-4eb3-8b5d-ea7afacf053e",
   "metadata": {},
   "source": [
    "モデルを作り、重回帰分析を行い、結果を表示します。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "2b5ca61a-1ee4-4423-ba53-0f47b61eab81",
   "metadata": {},
   "outputs": [],
   "source": [
    "model_3 = sm.OLS(y,X3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "b6c541a6-775b-4318-97e0-e4a631ac0efd",
   "metadata": {},
   "outputs": [],
   "source": [
    "result_3 = model_3.fit() "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "63fdc4ca-6d89-4eee-b25e-f83a0d522dc4",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                            OLS Regression Results                            \n",
      "==============================================================================\n",
      "Dep. Variable:                   rent   R-squared:                       0.760\n",
      "Model:                            OLS   Adj. R-squared:                  0.756\n",
      "Method:                 Least Squares   F-statistic:                     216.9\n",
      "Date:                Mon, 23 May 2022   Prob (F-statistic):           3.55e-43\n",
      "Time:                        10:56:39   Log-Likelihood:                -1479.6\n",
      "No. Observations:                 140   AIC:                             2965.\n",
      "Df Residuals:                     137   BIC:                             2974.\n",
      "Df Model:                           2                                         \n",
      "Covariance Type:            nonrobust                                         \n",
      "===============================================================================\n",
      "                  coef    std err          t      P>|t|      [0.025      0.975]\n",
      "-------------------------------------------------------------------------------\n",
      "const        4.757e+04   2910.072     16.348      0.000    4.18e+04    5.33e+04\n",
      "size         2177.3425    104.940     20.748      0.000    1969.831    2384.854\n",
      "proximity_B -8934.1895   1634.441     -5.466      0.000   -1.22e+04   -5702.194\n",
      "==============================================================================\n",
      "Omnibus:                        3.774   Durbin-Watson:                   1.557\n",
      "Prob(Omnibus):                  0.152   Jarque-Bera (JB):                3.881\n",
      "Skew:                           0.179   Prob(JB):                        0.144\n",
      "Kurtosis:                       3.733   Cond. No.                         102.\n",
      "==============================================================================\n",
      "\n",
      "Notes:\n",
      "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n"
     ]
    }
   ],
   "source": [
    "print(result_3.summary())"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a6e60776-16c8-43cc-8cc3-a9cdc4144122",
   "metadata": {},
   "source": [
    "この分析から、予測式は、家賃\\[円\\] = 47570 + 2177.3 x 広さ + (-8934.2) x 近さ  \n",
    "となります。どちらの説明変数もt値の絶対値が十分大きく、従ってp値が0に近く、十分に有意です。  \n",
    "広さのみを説明変数とした場合とくらべて、Adjusted R-squaredの値が大きいのがわかります。"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "755b8d41-83bd-46cf-a546-57dffed29ec2",
   "metadata": {},
   "source": [
    "以上から、２つめのモデル－広さと築年数－でも、３つめのモデル－広さと駅からの近さ－でも家賃をよく説明できると考えられます。"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "383b2750-ddd9-4b06-9ffc-59613f9f487d",
   "metadata": {},
   "source": [
    "### 重回帰分析の残差分析"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b6b501bb-ab6a-45cc-b5c1-06127353f001",
   "metadata": {},
   "source": [
    "単回帰分析と同様に、残差分析ができます。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "34d0408f-7e45-4c2e-803d-f7b5d5025ba3",
   "metadata": {},
   "outputs": [],
   "source": [
    "import lmdiag"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "4b30d803-c375-4c41-8cf0-95ef520a7d38",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<module 'matplotlib.pyplot' from 'E:\\\\PythonUSB\\\\PythonUSB\\\\Python\\\\python-3.9.9\\\\lib\\\\site-packages\\\\matplotlib\\\\pyplot.py'>"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "lmdiag.resid_fit(result_2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "860cafe5-a74a-448c-91ee-58011b136b04",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<module 'matplotlib.pyplot' from 'E:\\\\PythonUSB\\\\PythonUSB\\\\Python\\\\python-3.9.9\\\\lib\\\\site-packages\\\\matplotlib\\\\pyplot.py'>"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "lmdiag.resid_fit(result_3)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "bd9dff49-26da-4765-8cd1-bd66b352b702",
   "metadata": {},
   "source": [
    "予測値－残差プロットでは、モデル２－広さと築年数－もモデル３－広さと駅からの近さ－のどちらも単回帰分析のときより良さそうに見えます。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "01c9deae-fb5f-41c9-b43b-929b0aeec094",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<module 'matplotlib.pyplot' from 'E:\\\\PythonUSB\\\\PythonUSB\\\\Python\\\\python-3.9.9\\\\lib\\\\site-packages\\\\matplotlib\\\\pyplot.py'>"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "lmdiag.q_q(result_2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "id": "79b59f00-65c7-4e06-8e22-68a7523f2e3a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<module 'matplotlib.pyplot' from 'E:\\\\PythonUSB\\\\PythonUSB\\\\Python\\\\python-3.9.9\\\\lib\\\\site-packages\\\\matplotlib\\\\pyplot.py'>"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "lmdiag.q_q(result_3)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "42662e08-495c-45ed-86e8-db57135b5728",
   "metadata": {},
   "source": [
    "QQプロットでは、モデル２もモデル３もモデル１と同様に収まっているように見えます。"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "41ef827e-ea4d-42be-b55e-84ef553e3187",
   "metadata": {},
   "source": [
    "## （補足）分散分析"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "59b04eb0-7630-45b4-b94e-bbac5fd83662",
   "metadata": {
    "tags": []
   },
   "source": [
    "### 対象とするデータと必要とする理由"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3520ee57-18c6-40a6-89c0-63e0e9856cb6",
   "metadata": {},
   "source": [
    "分散分析は、次の例のように、通常３つ以上のグループ（水準あるいは級とよびます）間で平均値を比較する場合などに用いられます。"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5451a838-8f31-4a15-b887-4a4a9b4fb770",
   "metadata": {},
   "source": [
    "|メーカー|製品の耐久時間|平均値|\n",
    "|--|--|--|\n",
    "|A|1550, 1590, 1630, 1670, 1760|1640|\n",
    "|B|1530, 1590, 1600, 1640|1590|\n",
    "|C|1410, 1550, 1590, 1600, 1610, 1680, 1760|1600|\n",
    "|D|1440, 1480, 1550, 1610|1520|"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "078852b3-c039-4b15-b6f8-782cf5b37ef3",
   "metadata": {},
   "source": [
    "このように３つ以上の標本に対して、それらの間での平均値の差に有意性があることを検定するには、２標本の検定を繰り返してはいけません。"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8e3eea19-e7b1-4262-a8ca-c98d02cb46d3",
   "metadata": {},
   "source": [
    "なぜなら、検定を繰り返すことで第１種のエラー（$H_0$を棄却すべきでないのに棄却すること）の可能性が高まるからです。"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d502a4b4-30f3-4e55-9075-13a91ba28c1d",
   "metadata": {},
   "source": [
    "例えば５％水準で検定すると、１回の検定では間違って棄却はされない確率が９５％ですが、３つのグループがあったら３回検定することになりますので、間違って棄却はされない確率は$0.95^3 = 0.857$となり、第１種のエラーの確率は$1 - 0.857 = 0.143$に上がります。"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1a9fa08e-d5ad-49d4-8f3f-675c9facf024",
   "metadata": {},
   "source": [
    "### 基本的な考え方"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9f54d462-1684-4261-901e-04bf8bd99ff6",
   "metadata": {},
   "source": [
    "個々のデータの、全体の平均からの変動を、水準内でのばらつきと、水準の全体平均からの変動とに分解します。後者（水準間のばらつき）が前者（水準内のばらつうき）よりもはるかに大きければ、水準間に平均の有意な差があると判断します。"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "76d89fa8-ce42-4a8a-aa55-ea003879b2ac",
   "metadata": {},
   "source": [
    "### 分散分析（１元配置分散分析）のモデル"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "32abcfff-2081-41e7-9e71-3b116191430b",
   "metadata": {},
   "source": [
    "$i$番目の水準の$j$番目の確率変数を$x_{ij}$とします。これらは平均が$\\mu_0$、分散が$\\sigma^2$の正規分布を示すとします。  \n",
    "このとき、$x_{ij}$は次のように表せます。  \n",
    "$$x_{ij} = \\mu_i + \\epsilon_{ij}$$"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8f021e89-9086-4935-b63c-0d4b7f78bc85",
   "metadata": {},
   "source": [
    "ただし、$i = 1, 2, ..., k$で、$j = 1, 2, ..., n_i$"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "19759ce0-327f-466a-8755-3eb8ba5cc8ce",
   "metadata": {},
   "source": [
    "ここで$\\epsilon_{ij}$は誤差項で、平均値0, 分散$\\sigma^2$の独立正規分布をするものとします。"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1915565e-e8b9-4d65-a55c-d95747d23d07",
   "metadata": {},
   "source": [
    "さらに、平均値$\\mu_i$を$\\mu + \\alpha_i$と置き換えると、$x_{ij}$は、"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e7c01b4a-8d66-4b4d-8a31-8e3bf7eae8b6",
   "metadata": {},
   "source": [
    "$$x_{ij} = \\mu + \\alpha_i + \\epsilon_{ij} $$"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "73f02c9f-9d49-4ac0-8d18-8f0254fd0d75",
   "metadata": {},
   "source": [
    "(ただし、$\\sum \\alpha_i = 0$ )と表せます。"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3b84919a-eee5-4948-8df3-65c57226e3d0",
   "metadata": {},
   "source": [
    "このように、$x_{ij}$の変動は２つの効果の結果です。ひとつは、 $\\mu$があらわす、すべての変数に共通の効果です。もう１つは、$\\alpha_i$が表す、個々の水準$i$の母集団に固有の効果です。これを水準$i$への効果といいます。"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4c7454ed-579c-4c1c-9a95-5ac11c509998",
   "metadata": {},
   "source": [
    "### 水準内の分散と水準間の分散"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ff63de4e-4470-4291-a92d-c990cf420d6f",
   "metadata": {},
   "source": [
    "これらの変数の分散からいくつかの$\\chi^2$分布がつくれます。前提として、すべての水準が同じ母集団（分散が$\\sigma^2$）から出ているものとします。"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5cbcbde7-bfc3-45ad-a65e-52aeb36f9457",
   "metadata": {},
   "source": [
    "まず、$x_{ij}$は正規分布し、母分散が$\\sigma^2$ですから、全体の平均を$\\bar{x}$とすると、$$\\frac{\\sum\\sum(x_{ij} - \\bar{x})^2}{\\sigma^2} = \\frac{V}{\\sigma^2} $$は自由度$N - 1$の$\\chi^2$分布をします。"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e0dbad24-ee13-42d0-962d-4527c47d4bfc",
   "metadata": {},
   "source": [
    "同様に、各水準の平均を$\\bar{x_i}$とすると、**水準内の分散**に関しては、$$\\frac{\\sum\\sum(x_{ij} - \\bar{x_i})^2}{\\sigma^2} = \\frac{V_1}{\\sigma^2}$$は自由度$N - k$の$\\chi^2$分布をします。"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e56e4ee1-d1d6-4790-b52e-6416cad88d9b",
   "metadata": {},
   "source": [
    "同様に、**水準間の分散**に関しては、$$\\frac{\\sum n_i(\\bar{x_i} - \\bar{x})^2}{\\sigma^2} = \\frac{V_2}{\\sigma^2}$$は自由度$k - 1$の$\\chi^2$分布をします。"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "222a8d3e-5d49-4343-8570-e9a6df6ae360",
   "metadata": {},
   "source": [
    "そして、$\\chi^2$変数の期待値はその自由度に等しいので、$V/\\sigma^2$, $V_1/\\sigma^2$, $V_2/\\sigma^2$をそれぞれの自由度で割ったものの期待値は$1$になります。つまり、$V/(N - 1)$、$V_1/(N - k)$、$V_2/(k - 1)$の期待値はどれも$\\sigma^2$になります。すなわち、**これらはどれも母分散$\\sigma^2$の不偏推定量です**。"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "207ec28a-40db-4715-a851-749e3e44bd91",
   "metadata": {},
   "source": [
    "### 統計量と検定方法"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7a2aaad2-9c02-4f53-a14f-bffdbfda3b1d",
   "metadata": {},
   "source": [
    "帰無仮説$H_0$: $\\alpha_i = 0$ ($i = 1, 2, ..., k$)  \n",
    "対立仮説$H_1$: $\\alpha_i \\ne 0$  (少なくとも$i$の1つの値に対して)  \n",
    "とします。"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b03f5016-71a6-45b5-8c34-0d9957bd14d4",
   "metadata": {},
   "source": [
    "上の$V_1/(N - k)$と$V_2/(k - 1)$は互いに独立なので、$H_0$のもとでは、不偏分散の比$$F = \\frac{V_2/(k - 1)}{V_1/(N - k)}$$は自由度$k - 1$と$N - k$の$F$分布をします。"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "bf74e094-b5f9-4caa-a658-67d3abdeb463",
   "metadata": {},
   "source": [
    "そして、片側検定で有意水準を$\\alpha$とすれば、$F$を検定統計量として$H_0$の棄却域は、$F > F_\\alpha$で与えられます。"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2e61be51-aa76-4f80-9c42-cb30908cc398",
   "metadata": {},
   "source": [
    "#### 例題"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "991b556c-23ad-4e46-bf0c-e6f36b1578be",
   "metadata": {},
   "source": [
    "上の表で、メーカー間の製品寿命に差があるかどうかを調べます。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "9fe40a1c-976a-44a4-be0d-4a8f376251c8",
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "from scipy import stats"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "a7846c27-a36c-4330-a6b4-09ff0b94e18e",
   "metadata": {},
   "outputs": [],
   "source": [
    "maker_a = np.array([1550, 1590, 1630, 1670, 1760])\n",
    "maker_b = np.array([1530, 1590, 1600, 1640])\n",
    "maker_c = np.array([1410, 1550, 1590, 1600, 1610, 1680, 1760])\n",
    "maker_d = np.array([1440, 1480, 1550, 1610])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cdf97f93-3e0b-4b19-8093-868d59a87c0b",
   "metadata": {},
   "source": [
    "一元配置分散分析には、scipy.stats.f_oneway()などを用いることができます。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "a043508f-fe65-4ab2-ad5f-22dc1c84a29e",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "F_onewayResult(statistic=1.454222222222222, pvalue=0.26440210657974433)\n"
     ]
    }
   ],
   "source": [
    "print(stats.f_oneway(maker_a,maker_b, maker_c, maker_d))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1c3c6038-e6a0-4f5c-aaf6-1914f355f038",
   "metadata": {},
   "source": [
    "この$P-$値から、5%水準で有意でないことがわかります。つまり、メーカー間で製品寿命の平均に違いはありません。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "8a6ebc22-4ab4-4c91-b442-c2fe692403bb",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.9.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}