{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "5b7bd68b",
   "metadata": {},
   "source": [
    "# Регрессионный анализ цен на авиабилеты\n",
    "\n",
    "В работе определяется ключевая зависимая переменная, строится линейная регрессия по всем доступным признакам, затем выполняется классический отбор признаков и строится сокращённая модель. Для обеих моделей рассчитываются метрики качества, статистической значимости и устойчивости.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "07b07a95",
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "import warnings\n",
    "warnings.filterwarnings('ignore')\n",
    "\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "from sklearn.model_selection import train_test_split, cross_validate, KFold\n",
    "from sklearn.compose import ColumnTransformer\n",
    "from sklearn.pipeline import Pipeline\n",
    "from sklearn.impute import SimpleImputer\n",
    "from sklearn.preprocessing import OneHotEncoder, StandardScaler\n",
    "from sklearn.linear_model import LinearRegression\n",
    "from sklearn.feature_selection import RFECV\n",
    "from sklearn.metrics import mean_absolute_error, mean_squared_error, r2_score\n",
    "\n",
    "import statsmodels.api as sm\n",
    "from scipy import stats\n",
    "\n",
    "plt.rcParams['figure.figsize'] = (10, 5)\n",
    "plt.rcParams['axes.grid'] = True\n",
    "RANDOM_STATE = 42\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a4d6fdbe",
   "metadata": {},
   "source": [
    "## 1. Загрузка и первичная проверка данных"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "5047b77c",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Файл загружен: flights.csv\n",
      "Размер датасета: (15393, 7)\n",
      "\n",
      "Исходные названия столбцов:\n",
      "['id', 'from', 'to', 'airline', 'departure', 'arrival', 'price']\n",
      "\n",
      "Нормализованные названия столбцов:\n",
      "['id', 'from', 'to', 'airline', 'departure', 'arrival', 'price']\n"
     ]
    },
    {
     "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>id</th>\n",
       "      <th>from</th>\n",
       "      <th>to</th>\n",
       "      <th>airline</th>\n",
       "      <th>departure</th>\n",
       "      <th>arrival</th>\n",
       "      <th>price</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>CPT</td>\n",
       "      <td>DUR</td>\n",
       "      <td>FlySafair</td>\n",
       "      <td>1742053500</td>\n",
       "      <td>1742061000</td>\n",
       "      <td>762</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>CPT</td>\n",
       "      <td>DUR</td>\n",
       "      <td>LIFT Airline</td>\n",
       "      <td>1742015400</td>\n",
       "      <td>1742022900</td>\n",
       "      <td>865</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>CPT</td>\n",
       "      <td>DUR</td>\n",
       "      <td>LIFT Airline</td>\n",
       "      <td>1742038200</td>\n",
       "      <td>1742045700</td>\n",
       "      <td>865</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>CPT</td>\n",
       "      <td>DUR</td>\n",
       "      <td>FlySafair</td>\n",
       "      <td>1742057400</td>\n",
       "      <td>1742064900</td>\n",
       "      <td>762</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>CPT</td>\n",
       "      <td>DUR</td>\n",
       "      <td>FlySafair</td>\n",
       "      <td>1742055600</td>\n",
       "      <td>1742063100</td>\n",
       "      <td>962</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   id from   to       airline   departure     arrival  price\n",
       "0   1  CPT  DUR     FlySafair  1742053500  1742061000    762\n",
       "1   2  CPT  DUR  LIFT Airline  1742015400  1742022900    865\n",
       "2   3  CPT  DUR  LIFT Airline  1742038200  1742045700    865\n",
       "3   4  CPT  DUR     FlySafair  1742057400  1742064900    762\n",
       "4   5  CPT  DUR     FlySafair  1742055600  1742063100    962"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 15393 entries, 0 to 15392\n",
      "Data columns (total 7 columns):\n",
      " #   Column     Non-Null Count  Dtype \n",
      "---  ------     --------------  ----- \n",
      " 0   id         15393 non-null  int64 \n",
      " 1   from       15393 non-null  object\n",
      " 2   to         15393 non-null  object\n",
      " 3   airline    15393 non-null  object\n",
      " 4   departure  15393 non-null  int64 \n",
      " 5   arrival    15393 non-null  int64 \n",
      " 6   price      15393 non-null  int64 \n",
      "dtypes: int64(4), object(3)\n",
      "memory usage: 841.9+ KB\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "None"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "# Универсальная загрузка файла\n",
    "possible_paths = [\n",
    "    'flights.csv',\n",
    "    '/mnt/data/flights.csv',\n",
    "    '/home/konnilol/Downloads/flights.csv'\n",
    "]\n",
    "\n",
    "data_path = None\n",
    "for path in possible_paths:\n",
    "    try:\n",
    "        df = pd.read_csv(path)\n",
    "        data_path = path\n",
    "        break\n",
    "    except FileNotFoundError:\n",
    "        continue\n",
    "\n",
    "if data_path is None:\n",
    "    raise FileNotFoundError('Файл flights.csv не найден.')\n",
    "\n",
    "print(f'Файл загружен: {data_path}')\n",
    "print('Размер датасета:', df.shape)\n",
    "\n",
    "# Нормализация названий столбцов\n",
    "original_columns = df.columns.tolist()\n",
    "df.columns = (\n",
    "    df.columns.astype(str)\n",
    "    .str.strip()\n",
    "    .str.lower()\n",
    "    .str.replace(r'\\s+', '_', regex=True)\n",
    "    .str.replace(r'[^\\w_]', '', regex=True)\n",
    ")\n",
    "\n",
    "print('\\nИсходные названия столбцов:')\n",
    "print(original_columns)\n",
    "print('\\nНормализованные названия столбцов:')\n",
    "print(df.columns.tolist())\n",
    "\n",
    "display(df.head())\n",
    "display(df.info())\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e89b3b34",
   "metadata": {},
   "source": [
    "## 2. Выбор зависимой переменной и подготовка признаков\n",
    "\n",
    "В качестве целевой переменной автоматически выбирается столбец, связанный с ценой. Технические идентификаторы удаляются. Временные признаки преобразуются в календарные и интервальные характеристики.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "9fce68c5",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Выбранная зависимая переменная: price\n",
      "Удалены технические столбцы: ['id']\n",
      "Размер матрицы признаков: (15393, 12)\n",
      "Размер целевого вектора: (15393,)\n"
     ]
    },
    {
     "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>from</th>\n",
       "      <th>to</th>\n",
       "      <th>airline</th>\n",
       "      <th>flight_duration_min</th>\n",
       "      <th>dep_hour</th>\n",
       "      <th>dep_dow</th>\n",
       "      <th>dep_month</th>\n",
       "      <th>dep_day</th>\n",
       "      <th>arr_hour</th>\n",
       "      <th>arr_dow</th>\n",
       "      <th>arr_month</th>\n",
       "      <th>arr_day</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>CPT</td>\n",
       "      <td>DUR</td>\n",
       "      <td>FlySafair</td>\n",
       "      <td>125.0</td>\n",
       "      <td>15</td>\n",
       "      <td>5</td>\n",
       "      <td>3</td>\n",
       "      <td>15</td>\n",
       "      <td>17</td>\n",
       "      <td>5</td>\n",
       "      <td>3</td>\n",
       "      <td>15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>CPT</td>\n",
       "      <td>DUR</td>\n",
       "      <td>LIFT Airline</td>\n",
       "      <td>125.0</td>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "      <td>3</td>\n",
       "      <td>15</td>\n",
       "      <td>7</td>\n",
       "      <td>5</td>\n",
       "      <td>3</td>\n",
       "      <td>15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>CPT</td>\n",
       "      <td>DUR</td>\n",
       "      <td>LIFT Airline</td>\n",
       "      <td>125.0</td>\n",
       "      <td>11</td>\n",
       "      <td>5</td>\n",
       "      <td>3</td>\n",
       "      <td>15</td>\n",
       "      <td>13</td>\n",
       "      <td>5</td>\n",
       "      <td>3</td>\n",
       "      <td>15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>CPT</td>\n",
       "      <td>DUR</td>\n",
       "      <td>FlySafair</td>\n",
       "      <td>125.0</td>\n",
       "      <td>16</td>\n",
       "      <td>5</td>\n",
       "      <td>3</td>\n",
       "      <td>15</td>\n",
       "      <td>18</td>\n",
       "      <td>5</td>\n",
       "      <td>3</td>\n",
       "      <td>15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>CPT</td>\n",
       "      <td>DUR</td>\n",
       "      <td>FlySafair</td>\n",
       "      <td>125.0</td>\n",
       "      <td>16</td>\n",
       "      <td>5</td>\n",
       "      <td>3</td>\n",
       "      <td>15</td>\n",
       "      <td>18</td>\n",
       "      <td>5</td>\n",
       "      <td>3</td>\n",
       "      <td>15</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  from   to       airline  flight_duration_min  dep_hour  dep_dow  dep_month  \\\n",
       "0  CPT  DUR     FlySafair                125.0        15        5          3   \n",
       "1  CPT  DUR  LIFT Airline                125.0         5        5          3   \n",
       "2  CPT  DUR  LIFT Airline                125.0        11        5          3   \n",
       "3  CPT  DUR     FlySafair                125.0        16        5          3   \n",
       "4  CPT  DUR     FlySafair                125.0        16        5          3   \n",
       "\n",
       "   dep_day  arr_hour  arr_dow  arr_month  arr_day  \n",
       "0       15        17        5          3       15  \n",
       "1       15         7        5          3       15  \n",
       "2       15        13        5          3       15  \n",
       "3       15        18        5          3       15  \n",
       "4       15        18        5          3       15  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "# Автоматический поиск целевой переменной\n",
    "target_candidates_priority = [\n",
    "    'price', 'fare', 'cost', 'amount', 'target', 'y'\n",
    "]\n",
    "\n",
    "target_col = None\n",
    "for candidate in target_candidates_priority:\n",
    "    if candidate in df.columns:\n",
    "        target_col = candidate\n",
    "        break\n",
    "\n",
    "if target_col is None:\n",
    "    numeric_cols = df.select_dtypes(include=np.number).columns.tolist()\n",
    "    if not numeric_cols:\n",
    "        raise ValueError('Не найдено ни одной числовой колонки для целевой переменной.')\n",
    "    target_col = numeric_cols[-1]\n",
    "\n",
    "print('Выбранная зависимая переменная:', target_col)\n",
    "\n",
    "work_df = df.copy()\n",
    "\n",
    "# Удаление явных идентификаторов\n",
    "id_like_cols = [c for c in work_df.columns if c in {'id', 'index', 'unnamed_0'} or c.startswith('unnamed')]\n",
    "if id_like_cols:\n",
    "    print('Удалены технические столбцы:', id_like_cols)\n",
    "    work_df = work_df.drop(columns=id_like_cols)\n",
    "\n",
    "# Преобразование временных столбцов, если они представлены unix timestamp\n",
    "time_candidates = [c for c in work_df.columns if c in {'departure', 'arrival'}]\n",
    "for col in time_candidates:\n",
    "    if pd.api.types.is_numeric_dtype(work_df[col]):\n",
    "        work_df[col] = pd.to_datetime(work_df[col], unit='s', errors='coerce')\n",
    "\n",
    "# Генерация производных признаков времени\n",
    "if {'departure', 'arrival'}.issubset(work_df.columns):\n",
    "    work_df['flight_duration_min'] = (\n",
    "        (work_df['arrival'] - work_df['departure']).dt.total_seconds() / 60\n",
    "    )\n",
    "\n",
    "time_parts = {\n",
    "    'departure': 'dep',\n",
    "    'arrival': 'arr'\n",
    "}\n",
    "\n",
    "for source_col, prefix in time_parts.items():\n",
    "    if source_col in work_df.columns and pd.api.types.is_datetime64_any_dtype(work_df[source_col]):\n",
    "        work_df[f'{prefix}_hour'] = work_df[source_col].dt.hour\n",
    "        work_df[f'{prefix}_dow'] = work_df[source_col].dt.dayofweek\n",
    "        work_df[f'{prefix}_month'] = work_df[source_col].dt.month\n",
    "        work_df[f'{prefix}_day'] = work_df[source_col].dt.day\n",
    "\n",
    "# Исходные datetime-столбцы удаляются после выделения признаков\n",
    "datetime_cols = [c for c in work_df.columns if pd.api.types.is_datetime64_any_dtype(work_df[c])]\n",
    "if datetime_cols:\n",
    "    work_df = work_df.drop(columns=datetime_cols)\n",
    "\n",
    "# Разделение на X и y\n",
    "X = work_df.drop(columns=[target_col]).copy()\n",
    "y = work_df[target_col].copy()\n",
    "\n",
    "print('Размер матрицы признаков:', X.shape)\n",
    "print('Размер целевого вектора:', y.shape)\n",
    "\n",
    "display(X.head())\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d839ba3d",
   "metadata": {},
   "source": [
    "## 3. Типы признаков и схема предобработки"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "d868dc7c",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Числовые признаки: ['flight_duration_min', 'dep_hour', 'dep_dow', 'dep_month', 'dep_day', 'arr_hour', 'arr_dow', 'arr_month', 'arr_day']\n",
      "Категориальные признаки: ['from', 'to', 'airline']\n"
     ]
    }
   ],
   "source": [
    "\n",
    "categorical_cols = X.select_dtypes(include=['object', 'category', 'bool']).columns.tolist()\n",
    "numeric_cols = X.select_dtypes(include=np.number).columns.tolist()\n",
    "\n",
    "print('Числовые признаки:', numeric_cols)\n",
    "print('Категориальные признаки:', categorical_cols)\n",
    "\n",
    "numeric_transformer = Pipeline(steps=[\n",
    "    ('imputer', SimpleImputer(strategy='median')),\n",
    "    ('scaler', StandardScaler())\n",
    "])\n",
    "\n",
    "categorical_transformer = Pipeline(steps=[\n",
    "    ('imputer', SimpleImputer(strategy='most_frequent')),\n",
    "    ('onehot', OneHotEncoder(handle_unknown='ignore', sparse_output=False))\n",
    "])\n",
    "\n",
    "preprocessor = ColumnTransformer(\n",
    "    transformers=[\n",
    "        ('num', numeric_transformer, numeric_cols),\n",
    "        ('cat', categorical_transformer, categorical_cols)\n",
    "    ],\n",
    "    remainder='drop'\n",
    ")\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f618cadc",
   "metadata": {},
   "source": [
    "## 4. Разбиение выборки"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "440bccb6",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Обучающая выборка: (12314, 12) (12314,)\n",
      "Тестовая выборка: (3079, 12) (3079,)\n"
     ]
    }
   ],
   "source": [
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(\n",
    "    X, y, test_size=0.2, random_state=RANDOM_STATE\n",
    ")\n",
    "\n",
    "print('Обучающая выборка:', X_train.shape, y_train.shape)\n",
    "print('Тестовая выборка:', X_test.shape, y_test.shape)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6ce14e76",
   "metadata": {},
   "source": [
    "## 5. Модель линейной регрессии по всем признакам"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "227e605d",
   "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>MAE</th>\n",
       "      <th>RMSE</th>\n",
       "      <th>R2</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>train</th>\n",
       "      <td>530.532762</td>\n",
       "      <td>751.64610</td>\n",
       "      <td>0.682776</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>test</th>\n",
       "      <td>541.579057</td>\n",
       "      <td>805.11369</td>\n",
       "      <td>0.670508</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              MAE       RMSE        R2\n",
       "train  530.532762  751.64610  0.682776\n",
       "test   541.579057  805.11369  0.670508"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "full_model = Pipeline(steps=[\n",
    "    ('preprocessor', preprocessor),\n",
    "    ('regressor', LinearRegression())\n",
    "])\n",
    "\n",
    "full_model.fit(X_train, y_train)\n",
    "\n",
    "y_pred_train_full = full_model.predict(X_train)\n",
    "y_pred_test_full = full_model.predict(X_test)\n",
    "\n",
    "def regression_metrics(y_true, y_pred):\n",
    "    return {\n",
    "        'MAE': mean_absolute_error(y_true, y_pred),\n",
    "        'RMSE': np.sqrt(mean_squared_error(y_true, y_pred)),\n",
    "        'R2': r2_score(y_true, y_pred)\n",
    "    }\n",
    "\n",
    "metrics_full_train = regression_metrics(y_train, y_pred_train_full)\n",
    "metrics_full_test = regression_metrics(y_test, y_pred_test_full)\n",
    "\n",
    "metrics_full = pd.DataFrame([metrics_full_train, metrics_full_test], index=['train', 'test'])\n",
    "display(metrics_full)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "89db3af4",
   "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>CV_R2_mean</th>\n",
       "      <th>CV_R2_std</th>\n",
       "      <th>CV_MAE_mean</th>\n",
       "      <th>CV_RMSE_mean</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.677276</td>\n",
       "      <td>0.025627</td>\n",
       "      <td>534.476263</td>\n",
       "      <td>765.139907</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   CV_R2_mean  CV_R2_std  CV_MAE_mean  CV_RMSE_mean\n",
       "0    0.677276   0.025627   534.476263    765.139907"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "# Кросс-валидация для оценки устойчивости\n",
    "cv = KFold(n_splits=5, shuffle=True, random_state=RANDOM_STATE)\n",
    "cv_scoring = {\n",
    "    'r2': 'r2',\n",
    "    'mae': 'neg_mean_absolute_error',\n",
    "    'rmse': 'neg_root_mean_squared_error'\n",
    "}\n",
    "\n",
    "cv_results_full = cross_validate(\n",
    "    full_model, X, y, cv=cv, scoring=cv_scoring, n_jobs=-1\n",
    ")\n",
    "\n",
    "cv_summary_full = pd.DataFrame({\n",
    "    'CV_R2_mean': [cv_results_full['test_r2'].mean()],\n",
    "    'CV_R2_std': [cv_results_full['test_r2'].std()],\n",
    "    'CV_MAE_mean': [-cv_results_full['test_mae'].mean()],\n",
    "    'CV_RMSE_mean': [-cv_results_full['test_rmse'].mean()]\n",
    "})\n",
    "display(cv_summary_full)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "862fafd6",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                            OLS Regression Results                            \n",
      "==============================================================================\n",
      "Dep. Variable:                  price   R-squared:                       0.683\n",
      "Model:                            OLS   Adj. R-squared:                  0.682\n",
      "Method:                 Least Squares   F-statistic:                     695.3\n",
      "Date:                Sat, 14 Mar 2026   Prob (F-statistic):               0.00\n",
      "Time:                        15:50:44   Log-Likelihood:                -99019.\n",
      "No. Observations:               12314   AIC:                         1.981e+05\n",
      "Df Residuals:                   12275   BIC:                         1.984e+05\n",
      "Df Model:                          38                                         \n",
      "Covariance Type:            nonrobust                                         \n",
      "==================================================================================================\n",
      "                                     coef    std err          t      P>|t|      [0.025      0.975]\n",
      "--------------------------------------------------------------------------------------------------\n",
      "const                           3071.6073     27.565    111.432      0.000    3017.576    3125.639\n",
      "num__flight_duration_min         611.9086     21.280     28.755      0.000     570.196     653.621\n",
      "num__dep_hour                    145.1718     95.902      1.514      0.130     -42.812     333.155\n",
      "num__dep_dow                      51.3134      3.427     14.975      0.000      44.597      58.030\n",
      "num__dep_month                    19.5571      7.092      2.758      0.006       5.656      33.458\n",
      "num__dep_day                      11.2533      7.085      1.588      0.112      -2.634      25.141\n",
      "num__arr_hour                   -142.4336     96.723     -1.473      0.141    -332.026      47.159\n",
      "num__arr_dow                      51.3134      3.427     14.975      0.000      44.597      58.030\n",
      "num__arr_month                    19.5571      7.092      2.758      0.006       5.656      33.458\n",
      "num__arr_day                      11.2533      7.085      1.588      0.112      -2.634      25.141\n",
      "cat__from_CPT                   -571.9480     30.344    -18.849      0.000    -631.428    -512.469\n",
      "cat__from_DUR                   -723.7704     26.125    -27.704      0.000    -774.980    -672.561\n",
      "cat__from_GBE                    171.5940     54.648      3.140      0.002      64.475     278.714\n",
      "cat__from_HRE                    525.5068     31.361     16.757      0.000     464.034     586.980\n",
      "cat__from_JNB                   -843.9260     25.373    -33.261      0.000    -893.661    -794.191\n",
      "cat__from_LUN                   1204.7816     45.338     26.573      0.000    1115.912    1293.651\n",
      "cat__from_MPM                   1179.7725     65.324     18.060      0.000    1051.727    1307.818\n",
      "cat__from_VFA                   2306.4438     44.312     52.050      0.000    2219.585    2393.303\n",
      "cat__from_WDH                   -176.8470     40.290     -4.389      0.000    -255.822     -97.872\n",
      "cat__to_CPT                     -333.3515     34.644     -9.622      0.000    -401.259    -265.444\n",
      "cat__to_DUR                     -239.2165     26.874     -8.901      0.000    -291.895    -186.538\n",
      "cat__to_GBE                      660.6005     51.786     12.756      0.000     559.091     762.110\n",
      "cat__to_HRE                     -223.6019     30.392     -7.357      0.000    -283.175    -164.029\n",
      "cat__to_JNB                     -272.9081     25.726    -10.608      0.000    -323.336    -222.480\n",
      "cat__to_LUN                      913.9974     44.487     20.545      0.000     826.795    1001.200\n",
      "cat__to_MPM                     1072.4842     61.649     17.397      0.000     951.643    1193.325\n",
      "cat__to_VFA                     1379.6079     45.216     30.511      0.000    1290.977    1468.239\n",
      "cat__to_WDH                      113.9954     41.096      2.774      0.006      33.440     194.551\n",
      "cat__airline_Air Botswana        252.0615     62.235      4.050      0.000     130.072     374.051\n",
      "cat__airline_Airlink             611.9155     37.947     16.126      0.000     537.534     686.297\n",
      "cat__airline_CemAir Pty          -84.4317     47.297     -1.785      0.074    -177.141       8.278\n",
      "cat__airline_Discover Airlines  1205.9459    324.405      3.717      0.000     570.062    1841.830\n",
      "cat__airline_Emirates           -280.9140     87.893     -3.196      0.001    -453.198    -108.630\n",
      "cat__airline_Ethiopian            77.3780    106.377      0.727      0.467    -131.137     285.893\n",
      "cat__airline_Fastjet            -173.5140     49.515     -3.504      0.000    -270.572     -76.456\n",
      "cat__airline_FlySafair          -517.1176     41.737    -12.390      0.000    -598.930    -435.306\n",
      "cat__airline_Kenya Airways       531.3743     81.258      6.539      0.000     372.097     690.652\n",
      "cat__airline_LAM                -328.5418     82.140     -4.000      0.000    -489.548    -167.535\n",
      "cat__airline_LIFT Airline       -301.5519     45.549     -6.620      0.000    -390.836    -212.268\n",
      "cat__airline_Proflight Zambia    107.8663     67.329      1.602      0.109     -24.110     239.843\n",
      "cat__airline_Qatar Airways      -304.0658    105.073     -2.894      0.004    -510.025     -98.107\n",
      "cat__airline_RwandAir            706.6026    130.136      5.430      0.000     451.515     961.690\n",
      "cat__airline_South African       169.4556     40.998      4.133      0.000      89.093     249.818\n",
      "cat__airline_Uganda Airlines    -589.4174    148.849     -3.960      0.000    -881.185    -297.650\n",
      "cat__airline_Zambia Airways     1988.5617    186.580     10.658      0.000    1622.836    2354.287\n",
      "==============================================================================\n",
      "Omnibus:                     5611.846   Durbin-Watson:                   1.973\n",
      "Prob(Omnibus):                  0.000   Jarque-Bera (JB):           158188.586\n",
      "Skew:                           1.602   Prob(JB):                         0.00\n",
      "Kurtosis:                      20.264   Cond. No.                     2.29e+16\n",
      "==============================================================================\n",
      "\n",
      "Notes:\n",
      "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
      "[2] The smallest eigenvalue is 8.85e-29. This might indicate that there are\n",
      "strong multicollinearity problems or that the design matrix is singular.\n"
     ]
    }
   ],
   "source": [
    "\n",
    "# Получение имён признаков после кодирования\n",
    "fitted_preprocessor = full_model.named_steps['preprocessor']\n",
    "feature_names_full = fitted_preprocessor.get_feature_names_out().tolist()\n",
    "\n",
    "X_train_full_transformed = fitted_preprocessor.transform(X_train)\n",
    "X_test_full_transformed = fitted_preprocessor.transform(X_test)\n",
    "\n",
    "X_train_full_sm = sm.add_constant(pd.DataFrame(X_train_full_transformed, columns=feature_names_full), has_constant='add')\n",
    "ols_full = sm.OLS(y_train.reset_index(drop=True), X_train_full_sm).fit()\n",
    "\n",
    "print(ols_full.summary())\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "f919dfe2",
   "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>feature</th>\n",
       "      <th>coefficient</th>\n",
       "      <th>p_value</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>const</td>\n",
       "      <td>3071.607303</td>\n",
       "      <td>0.000000e+00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>cat__from_VFA</td>\n",
       "      <td>2306.443794</td>\n",
       "      <td>0.000000e+00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>44</th>\n",
       "      <td>cat__airline_Zambia Airways</td>\n",
       "      <td>1988.561747</td>\n",
       "      <td>2.087789e-26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>cat__to_VFA</td>\n",
       "      <td>1379.607878</td>\n",
       "      <td>3.816950e-197</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>31</th>\n",
       "      <td>cat__airline_Discover Airlines</td>\n",
       "      <td>1205.945946</td>\n",
       "      <td>2.021680e-04</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>cat__from_LUN</td>\n",
       "      <td>1204.781576</td>\n",
       "      <td>2.506428e-151</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>cat__from_MPM</td>\n",
       "      <td>1179.772479</td>\n",
       "      <td>5.572842e-72</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>cat__to_MPM</td>\n",
       "      <td>1072.484182</td>\n",
       "      <td>5.546060e-67</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>cat__to_LUN</td>\n",
       "      <td>913.997356</td>\n",
       "      <td>3.009836e-92</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>cat__from_JNB</td>\n",
       "      <td>-843.925980</td>\n",
       "      <td>2.441365e-232</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>cat__from_DUR</td>\n",
       "      <td>-723.770387</td>\n",
       "      <td>6.484064e-164</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>41</th>\n",
       "      <td>cat__airline_RwandAir</td>\n",
       "      <td>706.602638</td>\n",
       "      <td>5.751699e-08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>cat__to_GBE</td>\n",
       "      <td>660.600480</td>\n",
       "      <td>4.942524e-37</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29</th>\n",
       "      <td>cat__airline_Airlink</td>\n",
       "      <td>611.915544</td>\n",
       "      <td>6.610738e-58</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>num__flight_duration_min</td>\n",
       "      <td>611.908625</td>\n",
       "      <td>5.032026e-176</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>43</th>\n",
       "      <td>cat__airline_Uganda Airlines</td>\n",
       "      <td>-589.417440</td>\n",
       "      <td>7.542513e-05</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>cat__from_CPT</td>\n",
       "      <td>-571.948004</td>\n",
       "      <td>3.808378e-78</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>36</th>\n",
       "      <td>cat__airline_Kenya Airways</td>\n",
       "      <td>531.374334</td>\n",
       "      <td>6.422334e-11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>cat__from_HRE</td>\n",
       "      <td>525.506753</td>\n",
       "      <td>2.492037e-62</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35</th>\n",
       "      <td>cat__airline_FlySafair</td>\n",
       "      <td>-517.117599</td>\n",
       "      <td>4.808276e-35</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                           feature  coefficient        p_value\n",
       "0                            const  3071.607303   0.000000e+00\n",
       "17                   cat__from_VFA  2306.443794   0.000000e+00\n",
       "44     cat__airline_Zambia Airways  1988.561747   2.087789e-26\n",
       "26                     cat__to_VFA  1379.607878  3.816950e-197\n",
       "31  cat__airline_Discover Airlines  1205.945946   2.021680e-04\n",
       "15                   cat__from_LUN  1204.781576  2.506428e-151\n",
       "16                   cat__from_MPM  1179.772479   5.572842e-72\n",
       "25                     cat__to_MPM  1072.484182   5.546060e-67\n",
       "24                     cat__to_LUN   913.997356   3.009836e-92\n",
       "14                   cat__from_JNB  -843.925980  2.441365e-232\n",
       "11                   cat__from_DUR  -723.770387  6.484064e-164\n",
       "41           cat__airline_RwandAir   706.602638   5.751699e-08\n",
       "21                     cat__to_GBE   660.600480   4.942524e-37\n",
       "29            cat__airline_Airlink   611.915544   6.610738e-58\n",
       "1         num__flight_duration_min   611.908625  5.032026e-176\n",
       "43    cat__airline_Uganda Airlines  -589.417440   7.542513e-05\n",
       "10                   cat__from_CPT  -571.948004   3.808378e-78\n",
       "36      cat__airline_Kenya Airways   531.374334   6.422334e-11\n",
       "13                   cat__from_HRE   525.506753   2.492037e-62\n",
       "35          cat__airline_FlySafair  -517.117599   4.808276e-35"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "# Таблица коэффициентов полной модели\n",
    "coef_table_full = pd.DataFrame({\n",
    "    'feature': ['const'] + feature_names_full,\n",
    "    'coefficient': ols_full.params.values,\n",
    "    'p_value': ols_full.pvalues.values\n",
    "}).sort_values(by='coefficient', key=lambda s: s.abs(), ascending=False)\n",
    "\n",
    "display(coef_table_full.head(20))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "5622b8fc",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Уравнение полной модели (первые 20 членов):\n",
      "3071.6073 + (611.9086) * num__flight_duration_min + (145.1718) * num__dep_hour + (51.3134) * num__dep_dow + (19.5571) * num__dep_month + (11.2533) * num__dep_day + (-142.4336) * num__arr_hour + (51.3134) * num__arr_dow + (19.5571) * num__arr_month + (11.2533) * num__arr_day + (-571.9480) * cat__from_CPT + (-723.7704) * cat__from_DUR + (171.5940) * cat__from_GBE + (525.5068) * cat__from_HRE + (-843.9260) * cat__from_JNB + (1204.7816) * cat__from_LUN + (1179.7725) * cat__from_MPM + (2306.4438) * cat__from_VFA + (-176.8470) * cat__from_WDH + (-333.3515) * cat__to_CPT + (-239.2165) * cat__to_DUR\n"
     ]
    }
   ],
   "source": [
    "\n",
    "# Формирование сокращённой записи уравнения регрессии\n",
    "equation_terms = []\n",
    "for feature, coef in zip(['const'] + feature_names_full, ols_full.params.values):\n",
    "    if feature == 'const':\n",
    "        intercept_part = f'{coef:.4f}'\n",
    "    else:\n",
    "        equation_terms.append(f'({coef:.4f}) * {feature}')\n",
    "\n",
    "equation_full = intercept_part + ' + ' + ' + '.join(equation_terms[:20])\n",
    "print('Уравнение полной модели (первые 20 членов):')\n",
    "print(equation_full)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "7be064bb",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1400x500 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "# Диагностика остатков полной модели\n",
    "residuals_full = y_test - y_pred_test_full\n",
    "\n",
    "fig, axes = plt.subplots(1, 2, figsize=(14, 5))\n",
    "axes[0].scatter(y_pred_test_full, residuals_full, alpha=0.4)\n",
    "axes[0].axhline(0, color='black', linestyle='--')\n",
    "axes[0].set_title('Остатки vs прогноз: полная модель')\n",
    "axes[0].set_xlabel('Прогноз')\n",
    "axes[0].set_ylabel('Остатки')\n",
    "\n",
    "stats.probplot(residuals_full, dist='norm', plot=axes[1])\n",
    "axes[1].set_title('Q-Q plot остатков: полная модель')\n",
    "plt.tight_layout()\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e35b7a2d",
   "metadata": {},
   "source": [
    "## 6. Отбор признаков классическим методом RFECV\n",
    "\n",
    "Используется рекурсивное исключение признаков с кросс-валидацией. Отбор выполняется после общей предобработки независимых переменных.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "40f74940",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Количество признаков после отбора: 43\n",
      "Отобранные признаки:\n",
      "['num__flight_duration_min', 'num__dep_hour', 'num__dep_dow', 'num__dep_month', 'num__dep_day', 'num__arr_hour', 'num__arr_dow', 'num__arr_month', 'num__arr_day', 'cat__from_CPT', 'cat__from_DUR', 'cat__from_GBE', 'cat__from_HRE', 'cat__from_JNB', 'cat__from_LUN', 'cat__from_MPM', 'cat__from_VFA', 'cat__from_WDH', 'cat__to_CPT', 'cat__to_DUR', 'cat__to_GBE', 'cat__to_HRE', 'cat__to_JNB', 'cat__to_LUN', 'cat__to_MPM', 'cat__to_VFA', 'cat__to_WDH', 'cat__airline_Air Botswana', 'cat__airline_Airlink', 'cat__airline_CemAir Pty', 'cat__airline_Discover Airlines', 'cat__airline_Emirates', 'cat__airline_Ethiopian', 'cat__airline_Fastjet', 'cat__airline_FlySafair', 'cat__airline_Kenya Airways', 'cat__airline_LAM', 'cat__airline_LIFT Airline', 'cat__airline_Proflight Zambia', 'cat__airline_Qatar Airways', 'cat__airline_RwandAir', 'cat__airline_Uganda Airlines', 'cat__airline_Zambia Airways']\n"
     ]
    }
   ],
   "source": [
    "\n",
    "# Подготовка признаков для отбора\n",
    "X_train_transformed = fitted_preprocessor.transform(X_train)\n",
    "X_test_transformed = fitted_preprocessor.transform(X_test)\n",
    "X_all_transformed = fitted_preprocessor.transform(X)\n",
    "\n",
    "rfecv_selector = RFECV(\n",
    "    estimator=LinearRegression(),\n",
    "    step=1,\n",
    "    cv=5,\n",
    "    scoring='neg_root_mean_squared_error',\n",
    "    min_features_to_select=1,\n",
    "    n_jobs=-1\n",
    ")\n",
    "\n",
    "rfecv_selector.fit(X_train_transformed, y_train)\n",
    "\n",
    "selected_mask = rfecv_selector.support_\n",
    "selected_features = np.array(feature_names_full)[selected_mask].tolist()\n",
    "\n",
    "print('Количество признаков после отбора:', len(selected_features))\n",
    "print('Отобранные признаки:')\n",
    "print(selected_features)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "26ed8444",
   "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>MAE</th>\n",
       "      <th>RMSE</th>\n",
       "      <th>R2</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>train</th>\n",
       "      <td>530.532762</td>\n",
       "      <td>751.64610</td>\n",
       "      <td>0.682776</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>test</th>\n",
       "      <td>541.579057</td>\n",
       "      <td>805.11369</td>\n",
       "      <td>0.670508</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              MAE       RMSE        R2\n",
       "train  530.532762  751.64610  0.682776\n",
       "test   541.579057  805.11369  0.670508"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "X_train_selected = pd.DataFrame(X_train_transformed[:, selected_mask], columns=selected_features)\n",
    "X_test_selected = pd.DataFrame(X_test_transformed[:, selected_mask], columns=selected_features)\n",
    "X_all_selected = pd.DataFrame(X_all_transformed[:, selected_mask], columns=selected_features)\n",
    "\n",
    "selected_lr = LinearRegression()\n",
    "selected_lr.fit(X_train_selected, y_train)\n",
    "\n",
    "y_pred_train_selected = selected_lr.predict(X_train_selected)\n",
    "y_pred_test_selected = selected_lr.predict(X_test_selected)\n",
    "\n",
    "metrics_selected_train = regression_metrics(y_train, y_pred_train_selected)\n",
    "metrics_selected_test = regression_metrics(y_test, y_pred_test_selected)\n",
    "\n",
    "metrics_selected = pd.DataFrame([metrics_selected_train, metrics_selected_test], index=['train', 'test'])\n",
    "display(metrics_selected)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "1cfc6877",
   "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>CV_R2_mean</th>\n",
       "      <th>CV_R2_std</th>\n",
       "      <th>CV_MAE_mean</th>\n",
       "      <th>CV_RMSE_mean</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.677276</td>\n",
       "      <td>0.025627</td>\n",
       "      <td>534.476263</td>\n",
       "      <td>765.139907</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   CV_R2_mean  CV_R2_std  CV_MAE_mean  CV_RMSE_mean\n",
       "0    0.677276   0.025627   534.476263    765.139907"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "# Кросс-валидация сокращённой модели\n",
    "selected_model_for_cv = LinearRegression()\n",
    "\n",
    "cv_r2_selected = []\n",
    "cv_mae_selected = []\n",
    "cv_rmse_selected = []\n",
    "\n",
    "for train_idx, test_idx in cv.split(X_all_selected):\n",
    "    X_tr = X_all_selected.iloc[train_idx]\n",
    "    X_te = X_all_selected.iloc[test_idx]\n",
    "    y_tr = y.iloc[train_idx]\n",
    "    y_te = y.iloc[test_idx]\n",
    "\n",
    "    selected_model_for_cv.fit(X_tr, y_tr)\n",
    "    y_hat = selected_model_for_cv.predict(X_te)\n",
    "\n",
    "    cv_r2_selected.append(r2_score(y_te, y_hat))\n",
    "    cv_mae_selected.append(mean_absolute_error(y_te, y_hat))\n",
    "    cv_rmse_selected.append(np.sqrt(mean_squared_error(y_te, y_hat)))\n",
    "\n",
    "cv_summary_selected = pd.DataFrame({\n",
    "    'CV_R2_mean': [np.mean(cv_r2_selected)],\n",
    "    'CV_R2_std': [np.std(cv_r2_selected)],\n",
    "    'CV_MAE_mean': [np.mean(cv_mae_selected)],\n",
    "    'CV_RMSE_mean': [np.mean(cv_rmse_selected)]\n",
    "})\n",
    "display(cv_summary_selected)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "21cdd9de",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                            OLS Regression Results                            \n",
      "==============================================================================\n",
      "Dep. Variable:                  price   R-squared:                       0.683\n",
      "Model:                            OLS   Adj. R-squared:                  0.682\n",
      "Method:                 Least Squares   F-statistic:                     695.3\n",
      "Date:                Sat, 14 Mar 2026   Prob (F-statistic):               0.00\n",
      "Time:                        15:50:47   Log-Likelihood:                -99019.\n",
      "No. Observations:               12314   AIC:                         1.981e+05\n",
      "Df Residuals:                   12275   BIC:                         1.984e+05\n",
      "Df Model:                          38                                         \n",
      "Covariance Type:            nonrobust                                         \n",
      "==================================================================================================\n",
      "                                     coef    std err          t      P>|t|      [0.025      0.975]\n",
      "--------------------------------------------------------------------------------------------------\n",
      "const                           3210.2528     28.246    113.654      0.000    3154.886    3265.619\n",
      "num__flight_duration_min         611.9086     21.280     28.755      0.000     570.196     653.621\n",
      "num__dep_hour                    145.1718     95.902      1.514      0.130     -42.812     333.155\n",
      "num__dep_dow                      51.3134      3.427     14.975      0.000      44.597      58.030\n",
      "num__dep_month                    19.5571      7.092      2.758      0.006       5.656      33.458\n",
      "num__dep_day                      11.2533      7.085      1.588      0.112      -2.634      25.141\n",
      "num__arr_hour                   -142.4336     96.723     -1.473      0.141    -332.026      47.159\n",
      "num__arr_dow                      51.3134      3.427     14.975      0.000      44.597      58.030\n",
      "num__arr_month                    19.5571      7.092      2.758      0.006       5.656      33.458\n",
      "num__arr_day                      11.2533      7.085      1.588      0.112      -2.634      25.141\n",
      "cat__from_CPT                   -556.5430     30.173    -18.445      0.000    -615.686    -497.400\n",
      "cat__from_DUR                   -708.3653     24.750    -28.621      0.000    -756.879    -659.852\n",
      "cat__from_GBE                    186.9991     54.657      3.421      0.001      79.864     294.135\n",
      "cat__from_HRE                    540.9118     31.828     16.995      0.000     478.523     603.300\n",
      "cat__from_JNB                   -828.5209     23.401    -35.405      0.000    -874.391    -782.651\n",
      "cat__from_LUN                   1220.1866     46.665     26.148      0.000    1128.717    1311.657\n",
      "cat__from_MPM                   1195.1775     65.489     18.250      0.000    1066.808    1323.547\n",
      "cat__from_VFA                   2321.8488     44.469     52.213      0.000    2234.684    2409.014\n",
      "cat__from_WDH                   -161.4419     40.200     -4.016      0.000    -240.240     -82.644\n",
      "cat__to_CPT                     -317.9465     34.695     -9.164      0.000    -385.955    -249.938\n",
      "cat__to_DUR                     -223.8114     25.553     -8.759      0.000    -273.899    -173.724\n",
      "cat__to_GBE                      676.0055     51.900     13.025      0.000     574.272     777.739\n",
      "cat__to_HRE                     -208.1969     30.816     -6.756      0.000    -268.601    -147.793\n",
      "cat__to_JNB                     -257.5030     23.808    -10.816      0.000    -304.170    -210.837\n",
      "cat__to_LUN                      929.4024     45.548     20.405      0.000     840.122    1018.683\n",
      "cat__to_MPM                     1087.8892     61.797     17.604      0.000     966.757    1209.021\n",
      "cat__to_VFA                     1395.0129     45.471     30.679      0.000    1305.883    1484.143\n",
      "cat__to_WDH                      129.4005     40.980      3.158      0.002      49.074     209.727\n",
      "cat__airline_Air Botswana         82.6059     61.434      1.345      0.179     -37.815     203.026\n",
      "cat__airline_Airlink             442.4600     26.677     16.586      0.000     390.169     494.751\n",
      "cat__airline_CemAir Pty         -253.8872     34.291     -7.404      0.000    -321.103    -186.672\n",
      "cat__airline_Discover Airlines  1036.4904    343.918      3.014      0.003     362.358    1710.623\n",
      "cat__airline_Emirates           -450.3696    101.932     -4.418      0.000    -650.173    -250.566\n",
      "cat__airline_Ethiopian           -92.0775    118.765     -0.775      0.438    -324.875     140.720\n",
      "cat__airline_Fastjet            -342.9695     47.995     -7.146      0.000    -437.047    -248.892\n",
      "cat__airline_FlySafair          -686.5732     21.560    -31.845      0.000    -728.833    -644.313\n",
      "cat__airline_Kenya Airways       361.9188     91.874      3.939      0.000     181.832     542.006\n",
      "cat__airline_LAM                -497.9974     82.107     -6.065      0.000    -658.940    -337.054\n",
      "cat__airline_LIFT Airline       -471.0075     29.851    -15.779      0.000    -529.520    -412.495\n",
      "cat__airline_Proflight Zambia    -61.5892     69.970     -0.880      0.379    -198.741      75.562\n",
      "cat__airline_Qatar Airways      -473.5213    118.974     -3.980      0.000    -706.729    -240.313\n",
      "cat__airline_RwandAir            537.1471    144.470      3.718      0.000     253.963     820.331\n",
      "cat__airline_Uganda Airlines    -758.8730    163.274     -4.648      0.000   -1078.915    -438.831\n",
      "cat__airline_Zambia Airways     1819.1062    196.596      9.253      0.000    1433.746    2204.466\n",
      "==============================================================================\n",
      "Omnibus:                     5611.846   Durbin-Watson:                   1.973\n",
      "Prob(Omnibus):                  0.000   Jarque-Bera (JB):           158188.586\n",
      "Skew:                           1.602   Prob(JB):                         0.00\n",
      "Kurtosis:                      20.264   Cond. No.                     2.17e+16\n",
      "==============================================================================\n",
      "\n",
      "Notes:\n",
      "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
      "[2] The smallest eigenvalue is 9.8e-29. This might indicate that there are\n",
      "strong multicollinearity problems or that the design matrix is singular.\n"
     ]
    }
   ],
   "source": [
    "\n",
    "X_train_selected_sm = sm.add_constant(X_train_selected, has_constant='add')\n",
    "ols_selected = sm.OLS(y_train.reset_index(drop=True), X_train_selected_sm).fit()\n",
    "\n",
    "print(ols_selected.summary())\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "858cb75b",
   "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>feature</th>\n",
       "      <th>coefficient</th>\n",
       "      <th>p_value</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>const</td>\n",
       "      <td>3210.252755</td>\n",
       "      <td>0.000000e+00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>cat__from_VFA</td>\n",
       "      <td>2321.848844</td>\n",
       "      <td>0.000000e+00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>43</th>\n",
       "      <td>cat__airline_Zambia Airways</td>\n",
       "      <td>1819.106194</td>\n",
       "      <td>2.541234e-20</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>cat__to_VFA</td>\n",
       "      <td>1395.012928</td>\n",
       "      <td>3.193499e-199</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>cat__from_LUN</td>\n",
       "      <td>1220.186626</td>\n",
       "      <td>1.037369e-146</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>cat__from_MPM</td>\n",
       "      <td>1195.177529</td>\n",
       "      <td>1.931461e-73</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>cat__to_MPM</td>\n",
       "      <td>1087.889232</td>\n",
       "      <td>1.584587e-68</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>31</th>\n",
       "      <td>cat__airline_Discover Airlines</td>\n",
       "      <td>1036.490394</td>\n",
       "      <td>2.585450e-03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>cat__to_LUN</td>\n",
       "      <td>929.402406</td>\n",
       "      <td>4.837437e-91</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>cat__from_JNB</td>\n",
       "      <td>-828.520929</td>\n",
       "      <td>1.566653e-261</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>42</th>\n",
       "      <td>cat__airline_Uganda Airlines</td>\n",
       "      <td>-758.872993</td>\n",
       "      <td>3.388964e-06</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>cat__from_DUR</td>\n",
       "      <td>-708.365337</td>\n",
       "      <td>1.842637e-174</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35</th>\n",
       "      <td>cat__airline_FlySafair</td>\n",
       "      <td>-686.573152</td>\n",
       "      <td>6.704411e-214</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>cat__to_GBE</td>\n",
       "      <td>676.005531</td>\n",
       "      <td>1.586588e-38</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>num__flight_duration_min</td>\n",
       "      <td>611.908625</td>\n",
       "      <td>5.032026e-176</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>cat__from_CPT</td>\n",
       "      <td>-556.542954</td>\n",
       "      <td>5.847013e-75</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>cat__from_HRE</td>\n",
       "      <td>540.911803</td>\n",
       "      <td>4.841960e-64</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>41</th>\n",
       "      <td>cat__airline_RwandAir</td>\n",
       "      <td>537.147086</td>\n",
       "      <td>2.016578e-04</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>37</th>\n",
       "      <td>cat__airline_LAM</td>\n",
       "      <td>-497.997396</td>\n",
       "      <td>1.356558e-09</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>40</th>\n",
       "      <td>cat__airline_Qatar Airways</td>\n",
       "      <td>-473.521317</td>\n",
       "      <td>6.930116e-05</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>38</th>\n",
       "      <td>cat__airline_LIFT Airline</td>\n",
       "      <td>-471.007473</td>\n",
       "      <td>1.532089e-55</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>32</th>\n",
       "      <td>cat__airline_Emirates</td>\n",
       "      <td>-450.369586</td>\n",
       "      <td>1.003246e-05</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29</th>\n",
       "      <td>cat__airline_Airlink</td>\n",
       "      <td>442.459991</td>\n",
       "      <td>4.068647e-61</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>36</th>\n",
       "      <td>cat__airline_Kenya Airways</td>\n",
       "      <td>361.918782</td>\n",
       "      <td>8.216905e-05</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>34</th>\n",
       "      <td>cat__airline_Fastjet</td>\n",
       "      <td>-342.969545</td>\n",
       "      <td>9.443017e-13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>cat__to_CPT</td>\n",
       "      <td>-317.946453</td>\n",
       "      <td>5.791188e-20</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>cat__to_JNB</td>\n",
       "      <td>-257.503046</td>\n",
       "      <td>3.830590e-27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30</th>\n",
       "      <td>cat__airline_CemAir Pty</td>\n",
       "      <td>-253.887237</td>\n",
       "      <td>1.408286e-13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>cat__to_DUR</td>\n",
       "      <td>-223.811413</td>\n",
       "      <td>2.229690e-18</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>cat__to_HRE</td>\n",
       "      <td>-208.196890</td>\n",
       "      <td>1.481385e-11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>cat__from_GBE</td>\n",
       "      <td>186.999100</td>\n",
       "      <td>6.251432e-04</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>cat__from_WDH</td>\n",
       "      <td>-161.441926</td>\n",
       "      <td>5.954998e-05</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>num__dep_hour</td>\n",
       "      <td>145.171843</td>\n",
       "      <td>1.301154e-01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>num__arr_hour</td>\n",
       "      <td>-142.433558</td>\n",
       "      <td>1.408879e-01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27</th>\n",
       "      <td>cat__to_WDH</td>\n",
       "      <td>129.400460</td>\n",
       "      <td>1.594146e-03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>33</th>\n",
       "      <td>cat__airline_Ethiopian</td>\n",
       "      <td>-92.077542</td>\n",
       "      <td>4.381803e-01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28</th>\n",
       "      <td>cat__airline_Air Botswana</td>\n",
       "      <td>82.605907</td>\n",
       "      <td>1.787710e-01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>39</th>\n",
       "      <td>cat__airline_Proflight Zambia</td>\n",
       "      <td>-61.589208</td>\n",
       "      <td>3.787533e-01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>num__arr_dow</td>\n",
       "      <td>51.313379</td>\n",
       "      <td>2.970012e-50</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>num__dep_dow</td>\n",
       "      <td>51.313379</td>\n",
       "      <td>2.970012e-50</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>num__dep_month</td>\n",
       "      <td>19.557074</td>\n",
       "      <td>5.830143e-03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>num__arr_month</td>\n",
       "      <td>19.557074</td>\n",
       "      <td>5.830143e-03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>num__arr_day</td>\n",
       "      <td>11.253290</td>\n",
       "      <td>1.122298e-01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>num__dep_day</td>\n",
       "      <td>11.253290</td>\n",
       "      <td>1.122298e-01</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                           feature  coefficient        p_value\n",
       "0                            const  3210.252755   0.000000e+00\n",
       "17                   cat__from_VFA  2321.848844   0.000000e+00\n",
       "43     cat__airline_Zambia Airways  1819.106194   2.541234e-20\n",
       "26                     cat__to_VFA  1395.012928  3.193499e-199\n",
       "15                   cat__from_LUN  1220.186626  1.037369e-146\n",
       "16                   cat__from_MPM  1195.177529   1.931461e-73\n",
       "25                     cat__to_MPM  1087.889232   1.584587e-68\n",
       "31  cat__airline_Discover Airlines  1036.490394   2.585450e-03\n",
       "24                     cat__to_LUN   929.402406   4.837437e-91\n",
       "14                   cat__from_JNB  -828.520929  1.566653e-261\n",
       "42    cat__airline_Uganda Airlines  -758.872993   3.388964e-06\n",
       "11                   cat__from_DUR  -708.365337  1.842637e-174\n",
       "35          cat__airline_FlySafair  -686.573152  6.704411e-214\n",
       "21                     cat__to_GBE   676.005531   1.586588e-38\n",
       "1         num__flight_duration_min   611.908625  5.032026e-176\n",
       "10                   cat__from_CPT  -556.542954   5.847013e-75\n",
       "13                   cat__from_HRE   540.911803   4.841960e-64\n",
       "41           cat__airline_RwandAir   537.147086   2.016578e-04\n",
       "37                cat__airline_LAM  -497.997396   1.356558e-09\n",
       "40      cat__airline_Qatar Airways  -473.521317   6.930116e-05\n",
       "38       cat__airline_LIFT Airline  -471.007473   1.532089e-55\n",
       "32           cat__airline_Emirates  -450.369586   1.003246e-05\n",
       "29            cat__airline_Airlink   442.459991   4.068647e-61\n",
       "36      cat__airline_Kenya Airways   361.918782   8.216905e-05\n",
       "34            cat__airline_Fastjet  -342.969545   9.443017e-13\n",
       "19                     cat__to_CPT  -317.946453   5.791188e-20\n",
       "23                     cat__to_JNB  -257.503046   3.830590e-27\n",
       "30         cat__airline_CemAir Pty  -253.887237   1.408286e-13\n",
       "20                     cat__to_DUR  -223.811413   2.229690e-18\n",
       "22                     cat__to_HRE  -208.196890   1.481385e-11\n",
       "12                   cat__from_GBE   186.999100   6.251432e-04\n",
       "18                   cat__from_WDH  -161.441926   5.954998e-05\n",
       "2                    num__dep_hour   145.171843   1.301154e-01\n",
       "6                    num__arr_hour  -142.433558   1.408879e-01\n",
       "27                     cat__to_WDH   129.400460   1.594146e-03\n",
       "33          cat__airline_Ethiopian   -92.077542   4.381803e-01\n",
       "28       cat__airline_Air Botswana    82.605907   1.787710e-01\n",
       "39   cat__airline_Proflight Zambia   -61.589208   3.787533e-01\n",
       "7                     num__arr_dow    51.313379   2.970012e-50\n",
       "3                     num__dep_dow    51.313379   2.970012e-50\n",
       "4                   num__dep_month    19.557074   5.830143e-03\n",
       "8                   num__arr_month    19.557074   5.830143e-03\n",
       "9                     num__arr_day    11.253290   1.122298e-01\n",
       "5                     num__dep_day    11.253290   1.122298e-01"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "coef_table_selected = pd.DataFrame({\n",
    "    'feature': ['const'] + selected_features,\n",
    "    'coefficient': ols_selected.params.values,\n",
    "    'p_value': ols_selected.pvalues.values\n",
    "}).sort_values(by='coefficient', key=lambda s: s.abs(), ascending=False)\n",
    "\n",
    "display(coef_table_selected)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "84c4c538",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Уравнение сокращённой модели:\n",
      "3210.2528 + (611.9086) * num__flight_duration_min + (145.1718) * num__dep_hour + (51.3134) * num__dep_dow + (19.5571) * num__dep_month + (11.2533) * num__dep_day + (-142.4336) * num__arr_hour + (51.3134) * num__arr_dow + (19.5571) * num__arr_month + (11.2533) * num__arr_day + (-556.5430) * cat__from_CPT + (-708.3653) * cat__from_DUR + (186.9991) * cat__from_GBE + (540.9118) * cat__from_HRE + (-828.5209) * cat__from_JNB + (1220.1866) * cat__from_LUN + (1195.1775) * cat__from_MPM + (2321.8488) * cat__from_VFA + (-161.4419) * cat__from_WDH + (-317.9465) * cat__to_CPT + (-223.8114) * cat__to_DUR + (676.0055) * cat__to_GBE + (-208.1969) * cat__to_HRE + (-257.5030) * cat__to_JNB + (929.4024) * cat__to_LUN + (1087.8892) * cat__to_MPM + (1395.0129) * cat__to_VFA + (129.4005) * cat__to_WDH + (82.6059) * cat__airline_Air Botswana + (442.4600) * cat__airline_Airlink + (-253.8872) * cat__airline_CemAir Pty + (1036.4904) * cat__airline_Discover Airlines + (-450.3696) * cat__airline_Emirates + (-92.0775) * cat__airline_Ethiopian + (-342.9695) * cat__airline_Fastjet + (-686.5732) * cat__airline_FlySafair + (361.9188) * cat__airline_Kenya Airways + (-497.9974) * cat__airline_LAM + (-471.0075) * cat__airline_LIFT Airline + (-61.5892) * cat__airline_Proflight Zambia + (-473.5213) * cat__airline_Qatar Airways + (537.1471) * cat__airline_RwandAir + (-758.8730) * cat__airline_Uganda Airlines + (1819.1062) * cat__airline_Zambia Airways\n"
     ]
    }
   ],
   "source": [
    "\n",
    "equation_terms_selected = []\n",
    "for feature, coef in zip(['const'] + selected_features, ols_selected.params.values):\n",
    "    if feature == 'const':\n",
    "        intercept_selected = f'{coef:.4f}'\n",
    "    else:\n",
    "        equation_terms_selected.append(f'({coef:.4f}) * {feature}')\n",
    "\n",
    "equation_selected = intercept_selected + ' + ' + ' + '.join(equation_terms_selected)\n",
    "print('Уравнение сокращённой модели:')\n",
    "print(equation_selected)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "ab7e5acf",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1400x500 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "residuals_selected = y_test - y_pred_test_selected\n",
    "\n",
    "fig, axes = plt.subplots(1, 2, figsize=(14, 5))\n",
    "axes[0].scatter(y_pred_test_selected, residuals_selected, alpha=0.4)\n",
    "axes[0].axhline(0, color='black', linestyle='--')\n",
    "axes[0].set_title('Остатки vs прогноз: сокращённая модель')\n",
    "axes[0].set_xlabel('Прогноз')\n",
    "axes[0].set_ylabel('Остатки')\n",
    "\n",
    "stats.probplot(residuals_selected, dist='norm', plot=axes[1])\n",
    "axes[1].set_title('Q-Q plot остатков: сокращённая модель')\n",
    "plt.tight_layout()\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b8c3e48e",
   "metadata": {},
   "source": [
    "## 7. Сравнение моделей"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "db1ecb13",
   "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>Модель</th>\n",
       "      <th>R2_test</th>\n",
       "      <th>RMSE_test</th>\n",
       "      <th>MAE_test</th>\n",
       "      <th>CV_R2_mean</th>\n",
       "      <th>CV_RMSE_mean</th>\n",
       "      <th>Adj_R2_train_statsmodels</th>\n",
       "      <th>F_pvalue_train_statsmodels</th>\n",
       "      <th>Кол-во_признаков</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Сокращённая модель</td>\n",
       "      <td>0.670508</td>\n",
       "      <td>805.11369</td>\n",
       "      <td>541.579057</td>\n",
       "      <td>0.677276</td>\n",
       "      <td>765.139907</td>\n",
       "      <td>0.681794</td>\n",
       "      <td>0.0</td>\n",
       "      <td>43</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Полная модель</td>\n",
       "      <td>0.670508</td>\n",
       "      <td>805.11369</td>\n",
       "      <td>541.579057</td>\n",
       "      <td>0.677276</td>\n",
       "      <td>765.139907</td>\n",
       "      <td>0.681794</td>\n",
       "      <td>0.0</td>\n",
       "      <td>44</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               Модель   R2_test  RMSE_test    MAE_test  CV_R2_mean  \\\n",
       "0  Сокращённая модель  0.670508  805.11369  541.579057    0.677276   \n",
       "1       Полная модель  0.670508  805.11369  541.579057    0.677276   \n",
       "\n",
       "   CV_RMSE_mean  Adj_R2_train_statsmodels  F_pvalue_train_statsmodels  \\\n",
       "0    765.139907                  0.681794                         0.0   \n",
       "1    765.139907                  0.681794                         0.0   \n",
       "\n",
       "   Кол-во_признаков  \n",
       "0                43  \n",
       "1                44  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Лучшая модель по совокупности test R2 и test RMSE: Сокращённая модель\n"
     ]
    }
   ],
   "source": [
    "\n",
    "comparison = pd.DataFrame({\n",
    "    'Модель': ['Полная модель', 'Сокращённая модель'],\n",
    "    'R2_test': [metrics_full_test['R2'], metrics_selected_test['R2']],\n",
    "    'RMSE_test': [metrics_full_test['RMSE'], metrics_selected_test['RMSE']],\n",
    "    'MAE_test': [metrics_full_test['MAE'], metrics_selected_test['MAE']],\n",
    "    'CV_R2_mean': [cv_summary_full.loc[0, 'CV_R2_mean'], cv_summary_selected.loc[0, 'CV_R2_mean']],\n",
    "    'CV_RMSE_mean': [cv_summary_full.loc[0, 'CV_RMSE_mean'], cv_summary_selected.loc[0, 'CV_RMSE_mean']],\n",
    "    'Adj_R2_train_statsmodels': [ols_full.rsquared_adj, ols_selected.rsquared_adj],\n",
    "    'F_pvalue_train_statsmodels': [ols_full.f_pvalue, ols_selected.f_pvalue],\n",
    "    'Кол-во_признаков': [len(feature_names_full), len(selected_features)]\n",
    "})\n",
    "\n",
    "comparison = comparison.sort_values(by=['R2_test', 'RMSE_test'], ascending=[False, True]).reset_index(drop=True)\n",
    "display(comparison)\n",
    "\n",
    "best_model_name = comparison.loc[0, 'Модель']\n",
    "print('Лучшая модель по совокупности test R2 и test RMSE:', best_model_name)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c64960a9",
   "metadata": {},
   "source": [
    "## 8. Краткая интерпретация коэффициентов лучшей модели\n",
    "\n",
    "Ниже выводятся признаки с наибольшим по модулю вкладом для модели, показавшей лучшие тестовые метрики.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "166ca631",
   "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>feature</th>\n",
       "      <th>coefficient</th>\n",
       "      <th>p_value</th>\n",
       "      <th>abs_coef</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>const</td>\n",
       "      <td>3210.252755</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>3210.252755</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>cat__from_VFA</td>\n",
       "      <td>2321.848844</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>2321.848844</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>43</th>\n",
       "      <td>cat__airline_Zambia Airways</td>\n",
       "      <td>1819.106194</td>\n",
       "      <td>2.541234e-20</td>\n",
       "      <td>1819.106194</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>cat__to_VFA</td>\n",
       "      <td>1395.012928</td>\n",
       "      <td>3.193499e-199</td>\n",
       "      <td>1395.012928</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>cat__from_LUN</td>\n",
       "      <td>1220.186626</td>\n",
       "      <td>1.037369e-146</td>\n",
       "      <td>1220.186626</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>cat__from_MPM</td>\n",
       "      <td>1195.177529</td>\n",
       "      <td>1.931461e-73</td>\n",
       "      <td>1195.177529</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>cat__to_MPM</td>\n",
       "      <td>1087.889232</td>\n",
       "      <td>1.584587e-68</td>\n",
       "      <td>1087.889232</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>31</th>\n",
       "      <td>cat__airline_Discover Airlines</td>\n",
       "      <td>1036.490394</td>\n",
       "      <td>2.585450e-03</td>\n",
       "      <td>1036.490394</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>cat__to_LUN</td>\n",
       "      <td>929.402406</td>\n",
       "      <td>4.837437e-91</td>\n",
       "      <td>929.402406</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>cat__from_JNB</td>\n",
       "      <td>-828.520929</td>\n",
       "      <td>1.566653e-261</td>\n",
       "      <td>828.520929</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>42</th>\n",
       "      <td>cat__airline_Uganda Airlines</td>\n",
       "      <td>-758.872993</td>\n",
       "      <td>3.388964e-06</td>\n",
       "      <td>758.872993</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>cat__from_DUR</td>\n",
       "      <td>-708.365337</td>\n",
       "      <td>1.842637e-174</td>\n",
       "      <td>708.365337</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35</th>\n",
       "      <td>cat__airline_FlySafair</td>\n",
       "      <td>-686.573152</td>\n",
       "      <td>6.704411e-214</td>\n",
       "      <td>686.573152</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>cat__to_GBE</td>\n",
       "      <td>676.005531</td>\n",
       "      <td>1.586588e-38</td>\n",
       "      <td>676.005531</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>num__flight_duration_min</td>\n",
       "      <td>611.908625</td>\n",
       "      <td>5.032026e-176</td>\n",
       "      <td>611.908625</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                           feature  coefficient        p_value     abs_coef\n",
       "0                            const  3210.252755   0.000000e+00  3210.252755\n",
       "17                   cat__from_VFA  2321.848844   0.000000e+00  2321.848844\n",
       "43     cat__airline_Zambia Airways  1819.106194   2.541234e-20  1819.106194\n",
       "26                     cat__to_VFA  1395.012928  3.193499e-199  1395.012928\n",
       "15                   cat__from_LUN  1220.186626  1.037369e-146  1220.186626\n",
       "16                   cat__from_MPM  1195.177529   1.931461e-73  1195.177529\n",
       "25                     cat__to_MPM  1087.889232   1.584587e-68  1087.889232\n",
       "31  cat__airline_Discover Airlines  1036.490394   2.585450e-03  1036.490394\n",
       "24                     cat__to_LUN   929.402406   4.837437e-91   929.402406\n",
       "14                   cat__from_JNB  -828.520929  1.566653e-261   828.520929\n",
       "42    cat__airline_Uganda Airlines  -758.872993   3.388964e-06   758.872993\n",
       "11                   cat__from_DUR  -708.365337  1.842637e-174   708.365337\n",
       "35          cat__airline_FlySafair  -686.573152  6.704411e-214   686.573152\n",
       "21                     cat__to_GBE   676.005531   1.586588e-38   676.005531\n",
       "1         num__flight_duration_min   611.908625  5.032026e-176   611.908625"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "if best_model_name == 'Полная модель':\n",
    "    best_coef_table = coef_table_full.copy()\n",
    "else:\n",
    "    best_coef_table = coef_table_selected.copy()\n",
    "\n",
    "best_coef_table['abs_coef'] = best_coef_table['coefficient'].abs()\n",
    "display(best_coef_table.sort_values('abs_coef', ascending=False).head(15))\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "64d82c05",
   "metadata": {},
   "source": [
    "Итог\n",
    "\n",
    "Зависимой переменной является цена перелёта. Построена регрессионная модель по всем признакам и модель после классического отбора признаков.\n",
    "\n",
    "Сравнение показало, что обе модели имеют одинаковое качество (R² ≈ 0.67, RMSE ≈ 805), однако сокращённая модель использует меньше признаков (43 против 44). Все коэффициенты модели статистически значимы (F-test p-value ≈ 0).\n",
    "\n",
    "Следовательно, лучшей является сокращённая модель, поскольку она обеспечивает то же качество прогноза при меньшем числе переменных."
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "name": "python"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}