{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "7c898a5d",
   "metadata": {},
   "source": [
    "# Logistic regression: elasticity\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "5e660515",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-09-14T14:44:31.675138Z",
     "iopub.status.busy": "2026-09-14T14:44:31.675138Z",
     "iopub.status.idle": "2026-09-14T14:44:36.623377Z",
     "shell.execute_reply": "2026-09-14T14:44:36.621842Z"
    }
   },
   "outputs": [],
   "source": [
    "from pathlib import Path\n",
    "import sys\n",
    "\n",
    "PROJECT_ROOT = Path.cwd().resolve()\n",
    "while not (PROJECT_ROOT / 'data.py').exists():\n",
    "    if PROJECT_ROOT.parent == PROJECT_ROOT:\n",
    "        raise FileNotFoundError('Open this notebook from inside organized_uncertainty_analysis.')\n",
    "    PROJECT_ROOT = PROJECT_ROOT.parent\n",
    "if str(PROJECT_ROOT.parent) not in sys.path:\n",
    "    sys.path.insert(0, str(PROJECT_ROOT.parent))\n",
    "\n",
    "import numpy as np\n",
    "from organized_uncertainty_analysis.logistic.data_fit_gof import prepare_full_trim, fit_logistic_x\n",
    "from organized_uncertainty_analysis.logistic import elasticity as el\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "af2edcbc",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-09-14T14:44:36.632763Z",
     "iopub.status.busy": "2026-09-14T14:44:36.630744Z",
     "iopub.status.idle": "2026-09-14T14:44:45.485776Z",
     "shell.execute_reply": "2026-09-14T14:44:45.483765Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[trim] removed index=53, x_raw=2.481226, y=0\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>Panel</th>\n",
       "      <th>transform</th>\n",
       "      <th>theta0</th>\n",
       "      <th>theta1</th>\n",
       "      <th>x50</th>\n",
       "      <th>s50</th>\n",
       "      <th>E_x50_theta0</th>\n",
       "      <th>E_x50_theta1</th>\n",
       "      <th>E_s50_theta0</th>\n",
       "      <th>E_s50_theta1</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>FULL-RAW</td>\n",
       "      <td>raw</td>\n",
       "      <td>-11.683870</td>\n",
       "      <td>5.686964</td>\n",
       "      <td>2.054500</td>\n",
       "      <td>1.421741</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>FULL-LOG</td>\n",
       "      <td>log</td>\n",
       "      <td>-7.527408</td>\n",
       "      <td>10.738157</td>\n",
       "      <td>2.015760</td>\n",
       "      <td>1.331775</td>\n",
       "      <td>0.700996</td>\n",
       "      <td>-0.700996</td>\n",
       "      <td>-0.700996</td>\n",
       "      <td>1.700996</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>TRIM-RAW</td>\n",
       "      <td>raw</td>\n",
       "      <td>-26.022861</td>\n",
       "      <td>14.660490</td>\n",
       "      <td>1.775033</td>\n",
       "      <td>3.665123</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>TRIM-LOG</td>\n",
       "      <td>log</td>\n",
       "      <td>-14.173996</td>\n",
       "      <td>24.851636</td>\n",
       "      <td>1.768876</td>\n",
       "      <td>3.512347</td>\n",
       "      <td>0.570345</td>\n",
       "      <td>-0.570345</td>\n",
       "      <td>-0.570345</td>\n",
       "      <td>1.570345</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      Panel transform     theta0     theta1       x50       s50  E_x50_theta0  \\\n",
       "0  FULL-RAW       raw -11.683870   5.686964  2.054500  1.421741      1.000000   \n",
       "1  FULL-LOG       log  -7.527408  10.738157  2.015760  1.331775      0.700996   \n",
       "2  TRIM-RAW       raw -26.022861  14.660490  1.775033  3.665123      1.000000   \n",
       "3  TRIM-LOG       log -14.173996  24.851636  1.768876  3.512347      0.570345   \n",
       "\n",
       "   E_x50_theta1  E_s50_theta0  E_s50_theta1  \n",
       "0     -1.000000      0.000000      1.000000  \n",
       "1     -0.700996     -0.700996      1.700996  \n",
       "2     -1.000000      0.000000      1.000000  \n",
       "3     -0.570345     -0.570345      1.570345  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1800x756 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "data = prepare_full_trim(PROJECT_ROOT)\n",
    "P = {}\n",
    "for dataset, (x, y) in data.items():\n",
    "    for transform in ('raw', 'log'):\n",
    "        key = f'{dataset}-{transform.upper()}'\n",
    "        P[key] = {'b': fit_logistic_x(x, y, transform=transform, l2=1e-8), 'transform': transform}\n",
    "from IPython.display import display\n",
    "elasticity_table = el.elasticity_table_4panels(P, make_plots=False)\n",
    "elasticity_table.to_csv(PROJECT_ROOT / 'outputs' / 'logistic_elasticity.csv', index=False)\n",
    "fig, axes = el.plot_combined_elasticity(elasticity_table)\n",
    "fig.savefig(PROJECT_ROOT / 'outputs' / 'logistic_elasticity.png', dpi=600, bbox_inches='tight')\n",
    "fig.savefig(PROJECT_ROOT / 'outputs' / 'logistic_elasticity.pdf', bbox_inches='tight')\n",
    "display(elasticity_table)\n",
    "import matplotlib.pyplot as plt\n",
    "display(fig)\n",
    "plt.close(fig)\n"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.12.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}