{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "a7c40345",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "best_model\n",
      "lightgbm                                 0.70\n",
      "Naive                                    0.08\n",
      "HistoricAverage                          0.05\n",
      "Holt                                     0.05\n",
      "HoltWinters                              0.05\n",
      "SeasonalExponentialSmoothingOptimized    0.02\n",
      "CrostonSBA                               0.02\n",
      "CrostonOptimized                         0.02\n",
      "Name: proportion, dtype: float64\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "df = pd.read_csv(\"../metrics/best_models.csv\")\n",
    "\n",
    "# global win rate\n",
    "print(df[\"best_model\"].value_counts(normalize=True).round(2))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "05afbf96",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "cv_bin  best_model                           \n",
      "Low     CrostonOptimized                          0\n",
      "        CrostonSBA                                0\n",
      "        HistoricAverage                           1\n",
      "        Holt                                      0\n",
      "        HoltWinters                               1\n",
      "        Naive                                     1\n",
      "        SeasonalExponentialSmoothingOptimized     1\n",
      "        lightgbm                                  9\n",
      "Mid     CrostonOptimized                          1\n",
      "        CrostonSBA                                1\n",
      "        HistoricAverage                           1\n",
      "        Holt                                      1\n",
      "        HoltWinters                               1\n",
      "        Naive                                     0\n",
      "        SeasonalExponentialSmoothingOptimized     0\n",
      "        lightgbm                                  8\n",
      "High    CrostonOptimized                          0\n",
      "        CrostonSBA                                0\n",
      "        HistoricAverage                           0\n",
      "        Holt                                      1\n",
      "        HoltWinters                               0\n",
      "        Naive                                     2\n",
      "        SeasonalExponentialSmoothingOptimized     0\n",
      "        lightgbm                                 11\n",
      "dtype: int64\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\topra\\AppData\\Local\\Temp\\ipykernel_2576\\2721712874.py:9: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.\n",
      "  print(best.groupby([\"cv_bin\",\"best_model\"]).size())\n"
     ]
    }
   ],
   "source": [
    "full = pd.read_csv(\"../data/processed/train.csv\")\n",
    "\n",
    "vol = full.groupby(\"id\")[\"sales\"].agg([\"mean\",\"std\"]).reset_index()\n",
    "vol[\"cv\"] = vol[\"std\"] / (vol[\"mean\"] + 1e-9)\n",
    "\n",
    "best = df.merge(vol[[\"id\",\"cv\"]], on=\"id\")\n",
    "\n",
    "best[\"cv_bin\"] = pd.qcut(best[\"cv\"], 3, labels=[\"Low\",\"Mid\",\"High\"])\n",
    "print(best.groupby([\"cv_bin\",\"best_model\"]).size())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "e7c767a9",
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.read_csv(\"../data/processed/train.csv\")\n",
    "\n",
    "g = df.groupby(\"id\")[\"sales\"]\n",
    "summary = g.agg([\"mean\",\"std\",\"count\"])\n",
    "summary = summary.rename(columns={\"count\":\"T\"})\n",
    "\n",
    "summary[\"N\"] = g.apply(lambda x: (x>0).sum())\n",
    "summary[\"ADI\"] = summary[\"T\"] / summary[\"N\"].replace(0,1)\n",
    "summary[\"CV2\"] = (summary[\"std\"]/summary[\"mean\"].replace(0,1))**2\n",
    "\n",
    "summary.to_csv(\"../metrics/demand_profile.csv\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "6acec387",
   "metadata": {},
   "outputs": [],
   "source": [
    "summary[\"ADI_class\"] = np.where(summary[\"ADI\"] > 1.32, \"High\", \"Low\")\n",
    "summary[\"CV2_class\"] = np.where(summary[\"CV2\"] > 0.49, \"High\", \"Low\")\n",
    "\n",
    "summary[\"regime\"] = summary[\"ADI_class\"] + \"-\" + summary[\"CV2_class\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "04e8efeb",
   "metadata": {},
   "outputs": [],
   "source": [
    "best = pd.read_csv(\"../metrics/best_models.csv\")\n",
    "merged = best.merge(summary[[\"ADI\",\"CV2\",\"regime\"]], on=\"id\", how=\"left\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "db10f026",
   "metadata": {},
   "outputs": [],
   "source": [
    "merged.groupby(\"regime\")[\"best_model\"].value_counts(normalize=True).to_csv(\"../metrics/regime_model_performance.csv\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "53439790",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "best_model\n",
       "CrostonOptimized                          1\n",
       "CrostonSBA                                1\n",
       "HistoricAverage                           2\n",
       "Holt                                      2\n",
       "HoltWinters                               2\n",
       "Naive                                     3\n",
       "SeasonalExponentialSmoothingOptimized     1\n",
       "lightgbm                                 28\n",
       "dtype: int64"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "merged.groupby('best_model').size()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "dccd9376",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "regime     best_model                           \n",
       "High-High  lightgbm                                 17\n",
       "           Naive                                     2\n",
       "           Holt                                      1\n",
       "Low-High   lightgbm                                  6\n",
       "           HoltWinters                               2\n",
       "           CrostonOptimized                          1\n",
       "           CrostonSBA                                1\n",
       "           HistoricAverage                           1\n",
       "           Holt                                      1\n",
       "Low-Low    lightgbm                                  5\n",
       "           HistoricAverage                           1\n",
       "           Naive                                     1\n",
       "           SeasonalExponentialSmoothingOptimized     1\n",
       "Name: count, dtype: int64"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "merged.groupby('regime')['best_model'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "32eff5fe",
   "metadata": {},
   "outputs": [],
   "source": [
    "merged.to_csv('../metrics/best_by_sku.csv')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2c05c469",
   "metadata": {},
   "source": [
    "##  Key Insight\n",
    "Although classical literature suggests that intermittent & highly variable demand should be handled by Croston-type methods, our empirical evaluation on SKU-level series showed an asymmetry: LightGBM generalizes extremely well even under High-High ADI/CV² regimes, implying latent autocorrelation and structure that classical smoothing does not capture"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "e11acb96",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "best_model\n",
      "lightgbm                                 0.70\n",
      "Naive                                    0.08\n",
      "HistoricAverage                          0.05\n",
      "Holt                                     0.05\n",
      "HoltWinters                              0.05\n",
      "SeasonalExponentialSmoothingOptimized    0.02\n",
      "CrostonSBA                               0.02\n",
      "CrostonOptimized                         0.02\n",
      "Name: proportion, dtype: float64\n"
     ]
    }
   ],
   "source": [
    "print(best[\"best_model\"].value_counts(normalize=True).round(2))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "192189b7",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "model\n",
      "lightgbm                                 2.505754\n",
      "SeasonalExponentialSmoothingOptimized    3.206895\n",
      "AutoARIMA                                3.584182\n",
      "HistoricAverage                          3.584731\n",
      "CrostonClassic                           3.644770\n",
      "CrostonSBA                               3.656476\n",
      "CrostonOptimized                         3.785437\n",
      "SimpleExponentialSmoothingOptimized      3.792451\n",
      "Holt                                     3.812089\n",
      "HoltWinters                              3.913065\n",
      "DynamicOptimizedTheta                    3.958835\n",
      "Theta                                    3.960178\n",
      "DynamicTheta                             3.961312\n",
      "OptimizedTheta                           3.961813\n",
      "SeasonalNaive                            4.716071\n",
      "WindowAverage                            5.234375\n",
      "Naive                                    6.369643\n",
      "RandomWalkWithDrift                      6.437935\n",
      "Name: score, dtype: float64\n"
     ]
    }
   ],
   "source": [
    "m = pd.read_csv(\"../metrics/combined_metrics.csv\")\n",
    "print(m.groupby(\"model\")[\"score\"].mean().sort_values())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "223d1a5c",
   "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>id</th>\n",
       "      <th>best_model</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>FOODS_1_018_CA_1_validation</td>\n",
       "      <td>lightgbm</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>FOODS_1_085_CA_1_validation</td>\n",
       "      <td>lightgbm</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>FOODS_2_013_CA_1_validation</td>\n",
       "      <td>lightgbm</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>FOODS_2_019_CA_1_validation</td>\n",
       "      <td>SeasonalExponentialSmoothingOptimized</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>FOODS_2_030_CA_1_validation</td>\n",
       "      <td>lightgbm</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>FOODS_2_181_CA_1_validation</td>\n",
       "      <td>lightgbm</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>FOODS_2_197_CA_1_validation</td>\n",
       "      <td>HistoricAverage</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>HOBBIES_1_001_CA_1_validation</td>\n",
       "      <td>lightgbm</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>HOBBIES_1_002_CA_1_validation</td>\n",
       "      <td>lightgbm</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>HOBBIES_1_003_CA_1_validation</td>\n",
       "      <td>lightgbm</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>HOBBIES_1_004_CA_1_validation</td>\n",
       "      <td>lightgbm</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>HOBBIES_1_005_CA_1_validation</td>\n",
       "      <td>lightgbm</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>HOBBIES_1_006_CA_1_validation</td>\n",
       "      <td>lightgbm</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>HOBBIES_1_007_CA_1_validation</td>\n",
       "      <td>lightgbm</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>HOBBIES_1_008_CA_1_validation</td>\n",
       "      <td>lightgbm</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>HOBBIES_1_009_CA_1_validation</td>\n",
       "      <td>lightgbm</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>HOBBIES_1_010_CA_1_validation</td>\n",
       "      <td>lightgbm</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>HOBBIES_1_011_CA_1_validation</td>\n",
       "      <td>Naive</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>HOBBIES_1_012_CA_1_validation</td>\n",
       "      <td>Holt</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>HOBBIES_1_013_CA_1_validation</td>\n",
       "      <td>lightgbm</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>HOBBIES_1_014_CA_1_validation</td>\n",
       "      <td>lightgbm</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>HOBBIES_1_015_CA_1_validation</td>\n",
       "      <td>CrostonSBA</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>HOBBIES_1_016_CA_1_validation</td>\n",
       "      <td>lightgbm</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>HOBBIES_1_017_CA_1_validation</td>\n",
       "      <td>lightgbm</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>HOBBIES_1_018_CA_1_validation</td>\n",
       "      <td>Naive</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>HOBBIES_1_019_CA_1_validation</td>\n",
       "      <td>lightgbm</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>HOBBIES_1_020_CA_1_validation</td>\n",
       "      <td>lightgbm</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27</th>\n",
       "      <td>HOBBIES_1_021_CA_1_validation</td>\n",
       "      <td>lightgbm</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28</th>\n",
       "      <td>HOBBIES_1_022_CA_1_validation</td>\n",
       "      <td>lightgbm</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29</th>\n",
       "      <td>HOBBIES_1_103_CA_1_validation</td>\n",
       "      <td>HoltWinters</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30</th>\n",
       "      <td>HOBBIES_1_134_CA_1_validation</td>\n",
       "      <td>lightgbm</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>31</th>\n",
       "      <td>HOBBIES_1_147_CA_1_validation</td>\n",
       "      <td>lightgbm</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>32</th>\n",
       "      <td>HOBBIES_1_178_CA_1_validation</td>\n",
       "      <td>Holt</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>33</th>\n",
       "      <td>HOBBIES_1_254_CA_1_validation</td>\n",
       "      <td>HistoricAverage</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>34</th>\n",
       "      <td>HOBBIES_1_256_CA_1_validation</td>\n",
       "      <td>CrostonOptimized</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35</th>\n",
       "      <td>HOBBIES_1_268_CA_1_validation</td>\n",
       "      <td>lightgbm</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>36</th>\n",
       "      <td>HOBBIES_1_337_CA_1_validation</td>\n",
       "      <td>lightgbm</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>37</th>\n",
       "      <td>HOUSEHOLD_1_243_CA_1_validation</td>\n",
       "      <td>lightgbm</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>38</th>\n",
       "      <td>HOUSEHOLD_1_373_CA_1_validation</td>\n",
       "      <td>Naive</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>39</th>\n",
       "      <td>HOUSEHOLD_1_494_CA_1_validation</td>\n",
       "      <td>HoltWinters</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                 id                             best_model\n",
       "0       FOODS_1_018_CA_1_validation                               lightgbm\n",
       "1       FOODS_1_085_CA_1_validation                               lightgbm\n",
       "2       FOODS_2_013_CA_1_validation                               lightgbm\n",
       "3       FOODS_2_019_CA_1_validation  SeasonalExponentialSmoothingOptimized\n",
       "4       FOODS_2_030_CA_1_validation                               lightgbm\n",
       "5       FOODS_2_181_CA_1_validation                               lightgbm\n",
       "6       FOODS_2_197_CA_1_validation                        HistoricAverage\n",
       "7     HOBBIES_1_001_CA_1_validation                               lightgbm\n",
       "8     HOBBIES_1_002_CA_1_validation                               lightgbm\n",
       "9     HOBBIES_1_003_CA_1_validation                               lightgbm\n",
       "10    HOBBIES_1_004_CA_1_validation                               lightgbm\n",
       "11    HOBBIES_1_005_CA_1_validation                               lightgbm\n",
       "12    HOBBIES_1_006_CA_1_validation                               lightgbm\n",
       "13    HOBBIES_1_007_CA_1_validation                               lightgbm\n",
       "14    HOBBIES_1_008_CA_1_validation                               lightgbm\n",
       "15    HOBBIES_1_009_CA_1_validation                               lightgbm\n",
       "16    HOBBIES_1_010_CA_1_validation                               lightgbm\n",
       "17    HOBBIES_1_011_CA_1_validation                                  Naive\n",
       "18    HOBBIES_1_012_CA_1_validation                                   Holt\n",
       "19    HOBBIES_1_013_CA_1_validation                               lightgbm\n",
       "20    HOBBIES_1_014_CA_1_validation                               lightgbm\n",
       "21    HOBBIES_1_015_CA_1_validation                             CrostonSBA\n",
       "22    HOBBIES_1_016_CA_1_validation                               lightgbm\n",
       "23    HOBBIES_1_017_CA_1_validation                               lightgbm\n",
       "24    HOBBIES_1_018_CA_1_validation                                  Naive\n",
       "25    HOBBIES_1_019_CA_1_validation                               lightgbm\n",
       "26    HOBBIES_1_020_CA_1_validation                               lightgbm\n",
       "27    HOBBIES_1_021_CA_1_validation                               lightgbm\n",
       "28    HOBBIES_1_022_CA_1_validation                               lightgbm\n",
       "29    HOBBIES_1_103_CA_1_validation                            HoltWinters\n",
       "30    HOBBIES_1_134_CA_1_validation                               lightgbm\n",
       "31    HOBBIES_1_147_CA_1_validation                               lightgbm\n",
       "32    HOBBIES_1_178_CA_1_validation                                   Holt\n",
       "33    HOBBIES_1_254_CA_1_validation                        HistoricAverage\n",
       "34    HOBBIES_1_256_CA_1_validation                       CrostonOptimized\n",
       "35    HOBBIES_1_268_CA_1_validation                               lightgbm\n",
       "36    HOBBIES_1_337_CA_1_validation                               lightgbm\n",
       "37  HOUSEHOLD_1_243_CA_1_validation                               lightgbm\n",
       "38  HOUSEHOLD_1_373_CA_1_validation                                  Naive\n",
       "39  HOUSEHOLD_1_494_CA_1_validation                            HoltWinters"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "best[['id','best_model']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "8009c179",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\topra\\AppData\\Local\\Temp\\ipykernel_2576\\3706954313.py:75: UserWarning: set_ticklabels() should only be used with a fixed number of ticks, i.e. after set_ticks() or using a FixedLocator.\n",
      "  ax.set_xticklabels(model_scores.index, rotation=45, ha=\"right\")\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1600x1000 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 640x480 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "\n",
    "model_scores = m.groupby(\"model\")[\"score\"].mean().sort_values()\n",
    "\n",
    "fig, ax = plt.subplots(figsize=(16, 10))\n",
    "\n",
    "# Identify the best model explicitly\n",
    "best_model = model_scores.index[0]\n",
    "best_value = model_scores.iloc[0]\n",
    "\n",
    "# Tier boundaries (tunable later)\n",
    "tierB_upper = 4.0\n",
    "tierC_upper = 4.5\n",
    "\n",
    "colors = []\n",
    "for model, score in model_scores.items():\n",
    "    if model == best_model:\n",
    "        colors.append(\"#0047AB\")     # Deep Royal Blue → winner\n",
    "    elif score < tierB_upper:\n",
    "        colors.append(\"#888888\")     # Grey → Tier-B acceptable\n",
    "    else:\n",
    "        colors.append(\"#C43131\")     # Executive Red → weak tier\n",
    "\n",
    "bars = ax.bar(model_scores.index, model_scores.values, color=colors)\n",
    "\n",
    "# -------------------------------------------------\n",
    "# Threshold reference lines\n",
    "# -------------------------------------------------\n",
    "ax.axhline(tierB_upper, color=\"#666666\", linestyle=\"--\", linewidth=1)\n",
    "ax.text(\n",
    "    len(model_scores)-0.5,\n",
    "    tierB_upper + 0.02,\n",
    "    \"Tier-B Upper Boundary\",\n",
    "    color=\"#444444\",\n",
    "    fontsize=10,\n",
    "    ha=\"right\"\n",
    ")\n",
    "\n",
    "ax.axhline(tierC_upper, color=\"#444444\", linestyle=\":\", linewidth=1)\n",
    "ax.text(\n",
    "    len(model_scores)-0.5,\n",
    "    tierC_upper + 0.02,\n",
    "    \"Tier-C Avoid Zone\",\n",
    "    color=\"#444444\",\n",
    "    fontsize=10,\n",
    "    ha=\"right\"\n",
    ")\n",
    "\n",
    "# -------------------------------------------------\n",
    "# Value annotations\n",
    "# -------------------------------------------------\n",
    "for bar, value in zip(bars, model_scores.values):\n",
    "    ax.text(\n",
    "        bar.get_x() + bar.get_width()/2,\n",
    "        value + 0.02,\n",
    "        f\"{value:.2f}\",\n",
    "        ha=\"center\",\n",
    "        fontsize=9,\n",
    "        fontweight=\"bold\" if value == best_value else \"normal\"\n",
    "    )\n",
    "\n",
    "# -------------------------------------------------\n",
    "# Styling\n",
    "# -------------------------------------------------\n",
    "ax.set_title(\n",
    "    \"Portfolio-Level Model Performance (Lower Score = Lower Error Risk)\",\n",
    "    fontsize=16,\n",
    "    fontweight=\"bold\"\n",
    ")\n",
    "\n",
    "ax.set_ylabel(\"Mean Score (MAE + |Bias|)\", fontsize=12)\n",
    "ax.set_xlabel(\"Forecasting Models\", fontsize=12)\n",
    "\n",
    "ax.set_xticklabels(model_scores.index, rotation=45, ha=\"right\")\n",
    "\n",
    "ax.spines[\"top\"].set_visible(False)\n",
    "ax.spines[\"right\"].set_visible(False)\n",
    "ax.grid(axis=\"y\", alpha=0.25)\n",
    "plt.savefig(\"../docs/model_score_ranking.png\", dpi=300, bbox_inches=\"tight\")\n",
    "plt.show()\n",
    "plt.tight_layout()\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e24d1e03",
   "metadata": {},
   "source": [
    "LightGBM is the dominant model by portfolio-level score.\n",
    "Tier-B statistical alternatives remain valid and stable but inferior.\n",
    "Naive, Window-Average, and Random Walk variants systematically increase error-bias risk and are excluded from deployment"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "de0eb0b5",
   "metadata": {},
   "source": [
    "## Summary of Model Performance\n",
    "------------------------    \n",
    "LightGBM consistently outperforms statistical baselines across the portfolio.\n",
    "LightGBM not only wins more SKUs, it reduces overall error and reduces directional bias.\n",
    "\n",
    "lightgbm → 2.5 portfolio-level score (best)\n",
    "best statistical models → 3.2-4\n",
    "worst classical baselines → 4.7–6.3\n",
    "\n",
    "\n",
    "The models most commonly used in manual or heuristic planning introduce 40–120% higher combined loss-bias exposure compared to even basic smoothing and ML alternatives."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6cc7cdce",
   "metadata": {},
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "sandbox",
   "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": "3.11.14"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}