{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "078900c6-d71d-44f0-9e76-cc5ce66f0ca9",
   "metadata": {},
   "source": [
    "# Logistic regression: using general fit (for paper)\n",
    "\n",
    "Exploring different fitting techniques and using MLE asymptotic approx of parameter distribution to estimate model CI intervals.\n",
    "\n",
    "Using Logistic regression class and considering full data and trimmed."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e18cec5e",
   "metadata": {},
   "source": [
    "## Common"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "3b351af7",
   "metadata": {},
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "import scipy\n",
    "\n",
    "import pandas as pd\n",
    "import os\n",
    "\n",
    "# our libs\n",
    "import data_utils\n",
    "import logit_utils_gen\n",
    "\n",
    "np.set_printoptions(precision=16)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e0de7f02-942b-42ff-874c-3c0e2ac1e0dc",
   "metadata": {},
   "source": [
    "## Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "3373f87f",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "results_path = \"./results\"\n",
    "if not os.path.exists(results_path): os.makedirs(results_path)\n",
    "\n",
    "data_path = \"../../data/\"\n",
    "suv_filename = os.path.join(data_path, \"suv_percentilesSLOthenUWM.mat\")\n",
    "flags_filename = os.path.join(data_path,\"flags_combined.mat\")\n",
    "normal_range_filename = os.path.join(data_path, \"normal_range.mat\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "811d990d",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "\"\"\"\n",
    "Loading all data\n",
    "\"\"\"\n",
    "suv_dict = scipy.io.loadmat(suv_filename)\n",
    "flags_dict = scipy.io.loadmat(flags_filename)\n",
    "normal_range_dict = scipy.io.loadmat(normal_range_filename)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "4a292c31",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.microsoft.datawrangler.viewer.v0+json": {
       "columns": [
        {
         "name": "index",
         "rawType": "int64",
         "type": "integer"
        },
        {
         "name": "X",
         "rawType": "float64",
         "type": "float"
        },
        {
         "name": "Y",
         "rawType": "uint16",
         "type": "integer"
        },
        {
         "name": "scale",
         "rawType": "object",
         "type": "string"
        },
        {
         "name": "dataset",
         "rawType": "object",
         "type": "string"
        }
       ],
       "ref": "27f01a32-438c-4e00-b401-84dbabb206ef",
       "rows": [
        [
         "0",
         "2.105301856994629",
         "1",
         "plain",
         "FULL"
        ],
        [
         "1",
         "1.5363825559616089",
         "0",
         "plain",
         "FULL"
        ],
        [
         "2",
         "1.2744046449661255",
         "0",
         "plain",
         "FULL"
        ],
        [
         "3",
         "1.1879351139068604",
         "0",
         "plain",
         "FULL"
        ],
        [
         "4",
         "1.674635887145996",
         "1",
         "plain",
         "FULL"
        ],
        [
         "5",
         "0.9956096410751343",
         "0",
         "plain",
         "FULL"
        ],
        [
         "6",
         "1.299513339996338",
         "0",
         "plain",
         "FULL"
        ],
        [
         "7",
         "1.4487504959106445",
         "0",
         "plain",
         "FULL"
        ],
        [
         "8",
         "1.413230061531067",
         "0",
         "plain",
         "FULL"
        ],
        [
         "9",
         "0.9771649837493896",
         "0",
         "plain",
         "FULL"
        ],
        [
         "10",
         "4.151835918426514",
         "1",
         "plain",
         "FULL"
        ],
        [
         "11",
         "1.2182496786117554",
         "0",
         "plain",
         "FULL"
        ],
        [
         "12",
         "1.4161512851715088",
         "0",
         "plain",
         "FULL"
        ],
        [
         "13",
         "1.0392482280731201",
         "0",
         "plain",
         "FULL"
        ],
        [
         "14",
         "0.9517614245414734",
         "0",
         "plain",
         "FULL"
        ],
        [
         "15",
         "1.1691256761550903",
         "0",
         "plain",
         "FULL"
        ],
        [
         "16",
         "1.2764477729797363",
         "0",
         "plain",
         "FULL"
        ],
        [
         "17",
         "2.093014717102051",
         "1",
         "plain",
         "FULL"
        ],
        [
         "18",
         "1.4033281803131104",
         "0",
         "plain",
         "FULL"
        ],
        [
         "19",
         "1.3139079809188843",
         "0",
         "plain",
         "FULL"
        ],
        [
         "20",
         "1.2648413181304932",
         "0",
         "plain",
         "FULL"
        ],
        [
         "21",
         "1.0153511762619019",
         "0",
         "plain",
         "FULL"
        ],
        [
         "22",
         "0.8774824142456055",
         "0",
         "plain",
         "FULL"
        ],
        [
         "23",
         "1.4715447425842285",
         "0",
         "plain",
         "FULL"
        ],
        [
         "24",
         "1.3867568969726562",
         "0",
         "plain",
         "FULL"
        ],
        [
         "25",
         "1.485943078994751",
         "0",
         "plain",
         "FULL"
        ],
        [
         "26",
         "1.1236064434051514",
         "0",
         "plain",
         "FULL"
        ],
        [
         "27",
         "1.4543627500534058",
         "0",
         "plain",
         "FULL"
        ],
        [
         "28",
         "1.4773426055908203",
         "0",
         "plain",
         "FULL"
        ],
        [
         "29",
         "0.9777777791023254",
         "0",
         "plain",
         "FULL"
        ],
        [
         "30",
         "1.3661614656448364",
         "0",
         "plain",
         "FULL"
        ],
        [
         "31",
         "1.1513208150863647",
         "0",
         "plain",
         "FULL"
        ],
        [
         "32",
         "1.5016859769821167",
         "0",
         "plain",
         "FULL"
        ],
        [
         "33",
         "1.1443067789077759",
         "0",
         "plain",
         "FULL"
        ],
        [
         "34",
         "1.1909537315368652",
         "0",
         "plain",
         "FULL"
        ],
        [
         "35",
         "2.2312088012695312",
         "1",
         "plain",
         "FULL"
        ],
        [
         "36",
         "1.4084224700927734",
         "0",
         "plain",
         "FULL"
        ],
        [
         "37",
         "0.7728464007377625",
         "0",
         "plain",
         "FULL"
        ],
        [
         "38",
         "1.2335227727890015",
         "0",
         "plain",
         "FULL"
        ],
        [
         "39",
         "1.200500249862671",
         "0",
         "plain",
         "FULL"
        ],
        [
         "40",
         "1.5441060066223145",
         "0",
         "plain",
         "FULL"
        ],
        [
         "41",
         "1.4373862743377686",
         "0",
         "plain",
         "FULL"
        ],
        [
         "42",
         "1.4075382947921753",
         "0",
         "plain",
         "FULL"
        ],
        [
         "43",
         "1.3086644411087036",
         "0",
         "plain",
         "FULL"
        ],
        [
         "44",
         "1.4602000713348389",
         "0",
         "plain",
         "FULL"
        ],
        [
         "45",
         "1.299428939819336",
         "0",
         "plain",
         "FULL"
        ],
        [
         "46",
         "1.2612273693084717",
         "0",
         "plain",
         "FULL"
        ],
        [
         "47",
         "1.1733968257904053",
         "0",
         "plain",
         "FULL"
        ],
        [
         "48",
         "1.786200761795044",
         "0",
         "plain",
         "FULL"
        ],
        [
         "49",
         "1.2313168048858643",
         "0",
         "plain",
         "FULL"
        ]
       ],
       "shape": {
        "columns": 4,
        "rows": 230
       }
      },
      "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>X</th>\n",
       "      <th>Y</th>\n",
       "      <th>scale</th>\n",
       "      <th>dataset</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2.105302</td>\n",
       "      <td>1</td>\n",
       "      <td>plain</td>\n",
       "      <td>FULL</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1.536383</td>\n",
       "      <td>0</td>\n",
       "      <td>plain</td>\n",
       "      <td>FULL</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1.274405</td>\n",
       "      <td>0</td>\n",
       "      <td>plain</td>\n",
       "      <td>FULL</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1.187935</td>\n",
       "      <td>0</td>\n",
       "      <td>plain</td>\n",
       "      <td>FULL</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1.674636</td>\n",
       "      <td>1</td>\n",
       "      <td>plain</td>\n",
       "      <td>FULL</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>225</th>\n",
       "      <td>0.428051</td>\n",
       "      <td>0</td>\n",
       "      <td>log</td>\n",
       "      <td>TRIM</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>226</th>\n",
       "      <td>0.418426</td>\n",
       "      <td>0</td>\n",
       "      <td>log</td>\n",
       "      <td>TRIM</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>227</th>\n",
       "      <td>0.444583</td>\n",
       "      <td>0</td>\n",
       "      <td>log</td>\n",
       "      <td>TRIM</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>228</th>\n",
       "      <td>0.210767</td>\n",
       "      <td>0</td>\n",
       "      <td>log</td>\n",
       "      <td>TRIM</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>229</th>\n",
       "      <td>0.364238</td>\n",
       "      <td>0</td>\n",
       "      <td>log</td>\n",
       "      <td>TRIM</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>230 rows × 4 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "            X  Y  scale dataset\n",
       "0    2.105302  1  plain    FULL\n",
       "1    1.536383  0  plain    FULL\n",
       "2    1.274405  0  plain    FULL\n",
       "3    1.187935  0  plain    FULL\n",
       "4    1.674636  1  plain    FULL\n",
       "..        ... ..    ...     ...\n",
       "225  0.428051  0    log    TRIM\n",
       "226  0.418426  0    log    TRIM\n",
       "227  0.444583  0    log    TRIM\n",
       "228  0.210767  0    log    TRIM\n",
       "229  0.364238  0    log    TRIM\n",
       "\n",
       "[230 rows x 4 columns]"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "\"\"\"\n",
    "Data preparation\n",
    "\"\"\"\n",
    "# get data\n",
    "perc = 95\n",
    "organ = \"lung\"\n",
    "\n",
    "x0, y0 = data_utils.get_data(organ, perc, suv_dict, flags_dict)\n",
    "\n",
    "# values for trimmed dataset\n",
    "value_to_drop = 2.48    # target value\n",
    "tol = 1e-2              # tolerance: matches 2.481... too\n",
    "drop_mask = (np.abs(x0 - value_to_drop) <= tol)\n",
    "\n",
    "datasets = [\"FULL\", \"TRIM\"]\n",
    "scales = [\"plain\", \"log\"]\n",
    "\n",
    "lst = []\n",
    "for dataset in datasets:\n",
    "    for scale in scales:\n",
    "\n",
    "        if dataset == \"TRIM\":\n",
    "            y_tmp = y0[~drop_mask]\n",
    "            x_tmp = x0[~drop_mask]\n",
    "        else:\n",
    "            y_tmp = y0[:]\n",
    "            x_tmp = x0[:]\n",
    "        \n",
    "        if scale == \"log\":\n",
    "            # remove non-positive values for log scale\n",
    "            valid_mask = (x_tmp > 0)\n",
    "            y_tmp = y_tmp[valid_mask]\n",
    "            x_tmp = x_tmp[valid_mask]\n",
    "            x_tmp = np.log(x_tmp)\n",
    "\n",
    "        df_tmp = pd.DataFrame({\n",
    "            \"X\" : x_tmp,\n",
    "            \"Y\" : y_tmp ,\n",
    "            \"scale\" : scale,\n",
    "            \"dataset\": dataset})\n",
    "        lst.append(df_tmp)\n",
    "\n",
    "df_data = pd.concat(lst, ignore_index=True)\n",
    "df_data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "f27d9a03",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Index(['X', 'Y', 'scale', 'dataset'], dtype='object')\n"
     ]
    }
   ],
   "source": [
    "print(df_data.columns)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "35e9bb7b-5a50-4cae-9af4-261c11453dd4",
   "metadata": {},
   "source": [
    "## Fit\n",
    "\n",
    "Ref: logistic regression documentation\n",
    "* https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
    "* https://stats.stackexchange.com/questions/186830/what-is-scikit-learns-logisticregression-minimizing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "239ae789",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting: log, FULL\n",
      "Fitting: log, TRIM\n",
      "Fitting: plain, FULL\n",
      "Fitting: plain, TRIM\n"
     ]
    },
    {
     "data": {
      "application/vnd.microsoft.datawrangler.viewer.v0+json": {
       "columns": [
        {
         "name": "index",
         "rawType": "int64",
         "type": "integer"
        },
        {
         "name": "scale",
         "rawType": "object",
         "type": "string"
        },
        {
         "name": "dataset",
         "rawType": "object",
         "type": "string"
        },
        {
         "name": "beta",
         "rawType": "object",
         "type": "unknown"
        },
        {
         "name": "nllf",
         "rawType": "float64",
         "type": "float"
        },
        {
         "name": "jac_norm",
         "rawType": "float64",
         "type": "float"
        },
        {
         "name": "jac",
         "rawType": "object",
         "type": "unknown"
        },
        {
         "name": "cov",
         "rawType": "object",
         "type": "unknown"
        },
        {
         "name": "cov_evals",
         "rawType": "object",
         "type": "unknown"
        },
        {
         "name": "SE",
         "rawType": "object",
         "type": "unknown"
        },
        {
         "name": "LCL",
         "rawType": "object",
         "type": "unknown"
        },
        {
         "name": "UCL",
         "rawType": "object",
         "type": "unknown"
        },
        {
         "name": "LLF",
         "rawType": "float64",
         "type": "float"
        },
        {
         "name": "AIC",
         "rawType": "float64",
         "type": "float"
        },
        {
         "name": "BIC",
         "rawType": "float64",
         "type": "float"
        },
        {
         "name": "A",
         "rawType": "float64",
         "type": "float"
        },
        {
         "name": "chi2",
         "rawType": "float64",
         "type": "float"
        },
        {
         "name": "p-value(chi2)",
         "rawType": "float64",
         "type": "float"
        },
        {
         "name": "n",
         "rawType": "int64",
         "type": "integer"
        },
        {
         "name": "k",
         "rawType": "int64",
         "type": "integer"
        },
        {
         "name": "dof",
         "rawType": "int64",
         "type": "integer"
        },
        {
         "name": "pars",
         "rawType": "object",
         "type": "unknown"
        },
        {
         "name": "cost",
         "rawType": "float64",
         "type": "float"
        },
        {
         "name": "success",
         "rawType": "bool",
         "type": "boolean"
        }
       ],
       "ref": "e7eb2c6b-58b8-4bb2-9199-fd59f4db8988",
       "rows": [
        [
         "0",
         "log",
         "FULL",
         "[ -40.70878972735575  152.89575359163732 -186.37209570461724\n   75.72601427528205]",
         "4.602970217814242",
         "9.027086951330198e-05",
         "[ 8.1420683856878694e-05 -7.9943668704818909e-09 -3.0130919162990252e-05\n  2.4731149780170963e-05]",
         "[[ 8.3883805854039139e+01 -1.2065827516445837e-07 -8.2580113146009353e+00\n   1.0902563880598862e+01]\n [-1.2065827516444577e-07  4.0604781627682005e+00 -2.3482768628867510e-08\n  -2.5665712800966406e-08]\n [-8.2580113146007381e+00 -2.3482768628866736e-08  4.3606740310000234e+00\n  -1.9554958752592133e+00]\n [ 1.0902563880598656e+01 -2.5665712800966412e-08 -1.9554958752592095e+00\n   1.6443441115921698e+00]]",
         "[8.6175686581831073e+01 7.4059426536744171e-03 3.7057314721465335e+00\n 4.0604781627682041e+00]",
         "[9.158810285950853  2.01506281856626   2.088222696696888\n 1.2823198164234106]",
         "[-58.65972802905442   -3.949450583437119  10.979585296682927\n -14.878402904372432]",
         "[-22.757851425657083    3.9494505185141597  19.16526785113295\n  -9.851801590668634 ]",
         "-4.602970217814242",
         "17.205940435628484",
         "25.44771247781416",
         "0.9482758620689655",
         "8.316716016778969",
         "0.9999999999999134",
         "58",
         "4",
         "54",
         "[-4.0708789727355750e+01 -3.2461479341402061e-08  1.5072426573907938e+01\n -1.2365102247520532e+01]",
         "4.605007497171726",
         "True"
        ],
        [
         "1",
         "log",
         "TRIM",
         "[ -200.               1039.2599847864008 -1801.9074537027811\n  1041.4046274080492]",
         "1.7460229568910197",
         "0.0026576468930959263",
         "[ 2.6545272709629156e-03  1.0774747219714547e-07  1.1138498022276384e-04\n -6.4539527948865899e-05]",
         "[[ 3.5223072002646641e+03  5.6510650260542593e-04  3.7519578495074001e+02\n  -2.6307733467090708e+02]\n [ 5.6510650260505708e-04  4.9725933505851529e+01  5.3963315762392421e-05\n  -5.6477922331232095e-05]\n [ 3.7519578495062575e+02  5.3963315762415149e-05  7.3848116881359900e+01\n  -3.4292676039898353e+01]\n [-2.6307733467083528e+02 -5.6477922331251855e-05 -3.4292676039900186e+01\n   2.0813548748687520e+01]]",
         "[3.582409214949331e+03 4.257330045671801e-03 3.455539361542090e+01\n 4.972593350576528e+01]",
         "[59.349028637920135  7.051661754923553  8.593492705609279\n  4.562186838423819]",
         "[-316.32195864775974   -13.820997711899082  -72.73760394850915\n   23.295833608595615]",
         "[-83.67804135224027   13.821008429718276 -39.051731539705436\n  41.1792773967023  ]",
         "-1.7460229568910197",
         "11.49204591378204",
         "19.66425098512024",
         "0.9824561403508771",
         "2.5851621524449944",
         "1.0",
         "57",
         "4",
         "53",
         "[-2.000000000000000e+02  5.358909598118482e-06 -5.589466774410729e+01\n  3.223755550264896e+01]",
         "1.7901864307580302",
         "True"
        ],
        [
         "2",
         "plain",
         "FULL",
         "[-123.99037983249336   163.79100271566466   -71.44674345215762\n   10.388517613469324]",
         "4.766500995076863",
         "0.00024992369877456034",
         "[ 2.4830372529508968e-04 -5.1078370730206309e-07 -1.0651648090012622e-05\n  2.6332426905029127e-05]",
         "[[ 3.5618452493607731e+02 -7.2596318316022759e-06 -4.3972257183112040e+00\n   1.5801407388696280e+01]\n [-7.2596318316021370e-06  1.7437136990827204e+00 -1.3377424852495599e-07\n  -3.7757956852786992e-07]\n [-4.3972257183111081e+00 -1.3377424852495890e-07  2.2419957912691826e-01\n  -3.1693711550403902e-01]\n [ 1.5801407388696090e+01 -3.7757956852784784e-07 -3.1693711550404874e-01\n   7.8928386781204085e-01]]",
         "[3.5694014047196197e+02 5.8370496111149832e-04 2.5728420609316183e-01\n 1.7437136990826039e+00]",
         "[18.872851531659897   1.3204975195291813  0.4734971796398773\n  0.888416494563243 ]",
         "[-160.98048912011836     -2.5881284706184324    4.654574231634367\n  -14.539350233079611 ]",
         "[-87.00027054486836    2.588126689284912   6.510649069385271\n -11.056821567849049]",
         "-4.766500995076863",
         "17.53300199015373",
         "25.7747740323394",
         "0.9655172413793104",
         "9.164952239066789",
         "0.9999999999992063",
         "58",
         "4",
         "54",
         "[-1.2399037983249336e+02 -8.9066675997321292e-07  5.5826116505098193e+00\n -1.2798085900464329e+01]",
         "4.782069565923425",
         "True"
        ],
        [
         "3",
         "plain",
         "TRIM",
         "[-200.                300.1738053663387  -153.83046339703475\n   27.172339597628  ]",
         "2.5210340145493144",
         "0.0010641844799308764",
         "[ 1.0634668984410064e-03  6.2867824186796373e-06 -1.8057355772788108e-05\n  3.4075966829000270e-05]",
         "[[ 1.0714657996613662e+05 -1.8690296529443017e+02 -3.2127534629793927e+03\n   5.4151369572644171e+03]\n [-1.8690296531621391e+02  2.7611422539600396e+00  3.1674402466294098e+00\n  -8.7814992136047323e+00]\n [-3.2127534636210362e+03  3.1674402470959753e+00  1.0392314655379708e+02\n  -1.6617675301749378e+02]\n [ 5.4151369562329628e+03 -8.7814992106306651e+00 -1.6617675295413846e+02\n   2.7577605229208461e+02]]",
         "[1.0751694101221768e+05 1.0292755177410472e+01 1.8056209017215807e+00\n 9.1893967656789634e-04]",
         "[327.3325220110837      1.6616685150655168  10.194270280593756\n  16.60650632409131  ]",
         "[-841.5599541103887     -6.40020180348376   -10.951725021439106\n  -49.586138226246845]",
         "[4.4155995411038862e+02 1.1341908406137113e-01 2.9009080175822483e+01\n 1.5510170382264381e+01]",
         "-2.5210340145493144",
         "13.04206802909863",
         "21.21427310043683",
         "0.9649122807017544",
         "4.605672178037899",
         "1.0",
         "57",
         "4",
         "53",
         "[-200.                  -3.143391359711194    9.02867757719169\n  -17.03798392199123 ]",
         "2.5614157053734736",
         "True"
        ]
       ],
       "shape": {
        "columns": 23,
        "rows": 4
       }
      },
      "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>scale</th>\n",
       "      <th>dataset</th>\n",
       "      <th>beta</th>\n",
       "      <th>nllf</th>\n",
       "      <th>jac_norm</th>\n",
       "      <th>jac</th>\n",
       "      <th>cov</th>\n",
       "      <th>cov_evals</th>\n",
       "      <th>SE</th>\n",
       "      <th>LCL</th>\n",
       "      <th>...</th>\n",
       "      <th>BIC</th>\n",
       "      <th>A</th>\n",
       "      <th>chi2</th>\n",
       "      <th>p-value(chi2)</th>\n",
       "      <th>n</th>\n",
       "      <th>k</th>\n",
       "      <th>dof</th>\n",
       "      <th>pars</th>\n",
       "      <th>cost</th>\n",
       "      <th>success</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>log</td>\n",
       "      <td>FULL</td>\n",
       "      <td>[-40.70878972735575, 152.89575359163732, -186....</td>\n",
       "      <td>4.602970</td>\n",
       "      <td>0.000090</td>\n",
       "      <td>[8.14206838568787e-05, -7.994366870481891e-09,...</td>\n",
       "      <td>[[83.88380585403914, -1.2065827516445837e-07, ...</td>\n",
       "      <td>[86.17568658183107, 0.007405942653674417, 3.70...</td>\n",
       "      <td>[9.158810285950853, 2.01506281856626, 2.088222...</td>\n",
       "      <td>[-58.65972802905442, -3.949450583437119, 10.97...</td>\n",
       "      <td>...</td>\n",
       "      <td>25.447712</td>\n",
       "      <td>0.948276</td>\n",
       "      <td>8.316716</td>\n",
       "      <td>1.0</td>\n",
       "      <td>58</td>\n",
       "      <td>4</td>\n",
       "      <td>54</td>\n",
       "      <td>[-40.70878972735575, -3.246147934140206e-08, 1...</td>\n",
       "      <td>4.605007</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>log</td>\n",
       "      <td>TRIM</td>\n",
       "      <td>[-200.0, 1039.2599847864008, -1801.90745370278...</td>\n",
       "      <td>1.746023</td>\n",
       "      <td>0.002658</td>\n",
       "      <td>[0.0026545272709629156, 1.0774747219714547e-07...</td>\n",
       "      <td>[[3522.307200264664, 0.0005651065026054259, 37...</td>\n",
       "      <td>[3582.409214949331, 0.004257330045671801, 34.5...</td>\n",
       "      <td>[59.349028637920135, 7.051661754923553, 8.5934...</td>\n",
       "      <td>[-316.32195864775974, -13.820997711899082, -72...</td>\n",
       "      <td>...</td>\n",
       "      <td>19.664251</td>\n",
       "      <td>0.982456</td>\n",
       "      <td>2.585162</td>\n",
       "      <td>1.0</td>\n",
       "      <td>57</td>\n",
       "      <td>4</td>\n",
       "      <td>53</td>\n",
       "      <td>[-200.0, 5.358909598118482e-06, -55.8946677441...</td>\n",
       "      <td>1.790186</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>plain</td>\n",
       "      <td>FULL</td>\n",
       "      <td>[-123.99037983249336, 163.79100271566466, -71....</td>\n",
       "      <td>4.766501</td>\n",
       "      <td>0.000250</td>\n",
       "      <td>[0.0002483037252950897, -5.107837073020631e-07...</td>\n",
       "      <td>[[356.1845249360773, -7.259631831602276e-06, -...</td>\n",
       "      <td>[356.94014047196197, 0.0005837049611114983, 0....</td>\n",
       "      <td>[18.872851531659897, 1.3204975195291813, 0.473...</td>\n",
       "      <td>[-160.98048912011836, -2.5881284706184324, 4.6...</td>\n",
       "      <td>...</td>\n",
       "      <td>25.774774</td>\n",
       "      <td>0.965517</td>\n",
       "      <td>9.164952</td>\n",
       "      <td>1.0</td>\n",
       "      <td>58</td>\n",
       "      <td>4</td>\n",
       "      <td>54</td>\n",
       "      <td>[-123.99037983249336, -8.906667599732129e-07, ...</td>\n",
       "      <td>4.782070</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>plain</td>\n",
       "      <td>TRIM</td>\n",
       "      <td>[-200.0, 300.1738053663387, -153.8304633970347...</td>\n",
       "      <td>2.521034</td>\n",
       "      <td>0.001064</td>\n",
       "      <td>[0.0010634668984410064, 6.286782418679637e-06,...</td>\n",
       "      <td>[[107146.57996613662, -186.90296529443017, -32...</td>\n",
       "      <td>[107516.94101221768, 10.292755177410472, 1.805...</td>\n",
       "      <td>[327.3325220110837, 1.6616685150655168, 10.194...</td>\n",
       "      <td>[-841.5599541103887, -6.40020180348376, -10.95...</td>\n",
       "      <td>...</td>\n",
       "      <td>21.214273</td>\n",
       "      <td>0.964912</td>\n",
       "      <td>4.605672</td>\n",
       "      <td>1.0</td>\n",
       "      <td>57</td>\n",
       "      <td>4</td>\n",
       "      <td>53</td>\n",
       "      <td>[-200.0, -3.143391359711194, 9.02867757719169,...</td>\n",
       "      <td>2.561416</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>4 rows × 23 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   scale dataset                                               beta      nllf  \\\n",
       "0    log    FULL  [-40.70878972735575, 152.89575359163732, -186....  4.602970   \n",
       "1    log    TRIM  [-200.0, 1039.2599847864008, -1801.90745370278...  1.746023   \n",
       "2  plain    FULL  [-123.99037983249336, 163.79100271566466, -71....  4.766501   \n",
       "3  plain    TRIM  [-200.0, 300.1738053663387, -153.8304633970347...  2.521034   \n",
       "\n",
       "   jac_norm                                                jac  \\\n",
       "0  0.000090  [8.14206838568787e-05, -7.994366870481891e-09,...   \n",
       "1  0.002658  [0.0026545272709629156, 1.0774747219714547e-07...   \n",
       "2  0.000250  [0.0002483037252950897, -5.107837073020631e-07...   \n",
       "3  0.001064  [0.0010634668984410064, 6.286782418679637e-06,...   \n",
       "\n",
       "                                                 cov  \\\n",
       "0  [[83.88380585403914, -1.2065827516445837e-07, ...   \n",
       "1  [[3522.307200264664, 0.0005651065026054259, 37...   \n",
       "2  [[356.1845249360773, -7.259631831602276e-06, -...   \n",
       "3  [[107146.57996613662, -186.90296529443017, -32...   \n",
       "\n",
       "                                           cov_evals  \\\n",
       "0  [86.17568658183107, 0.007405942653674417, 3.70...   \n",
       "1  [3582.409214949331, 0.004257330045671801, 34.5...   \n",
       "2  [356.94014047196197, 0.0005837049611114983, 0....   \n",
       "3  [107516.94101221768, 10.292755177410472, 1.805...   \n",
       "\n",
       "                                                  SE  \\\n",
       "0  [9.158810285950853, 2.01506281856626, 2.088222...   \n",
       "1  [59.349028637920135, 7.051661754923553, 8.5934...   \n",
       "2  [18.872851531659897, 1.3204975195291813, 0.473...   \n",
       "3  [327.3325220110837, 1.6616685150655168, 10.194...   \n",
       "\n",
       "                                                 LCL  ...        BIC  \\\n",
       "0  [-58.65972802905442, -3.949450583437119, 10.97...  ...  25.447712   \n",
       "1  [-316.32195864775974, -13.820997711899082, -72...  ...  19.664251   \n",
       "2  [-160.98048912011836, -2.5881284706184324, 4.6...  ...  25.774774   \n",
       "3  [-841.5599541103887, -6.40020180348376, -10.95...  ...  21.214273   \n",
       "\n",
       "          A      chi2  p-value(chi2)   n  k  dof  \\\n",
       "0  0.948276  8.316716            1.0  58  4   54   \n",
       "1  0.982456  2.585162            1.0  57  4   53   \n",
       "2  0.965517  9.164952            1.0  58  4   54   \n",
       "3  0.964912  4.605672            1.0  57  4   53   \n",
       "\n",
       "                                                pars      cost  success  \n",
       "0  [-40.70878972735575, -3.246147934140206e-08, 1...  4.605007     True  \n",
       "1  [-200.0, 5.358909598118482e-06, -55.8946677441...  1.790186     True  \n",
       "2  [-123.99037983249336, -8.906667599732129e-07, ...  4.782070     True  \n",
       "3  [-200.0, -3.143391359711194, 9.02867757719169,...  2.561416     True  \n",
       "\n",
       "[4 rows x 23 columns]"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# defining two-sided confidence intervals\n",
    "alpha = 0.05   # significance level\n",
    "probs = [alpha/2, 1 - alpha/2]\n",
    "\n",
    "# fits\n",
    "lg = logit_utils_gen.LogisticPolyRegression(3, mono = True, lam = (0, 1e-6))\n",
    "\n",
    "lst = []\n",
    "for (scale, dataset), g in df_data.groupby(['scale', 'dataset']):\n",
    "    \n",
    "    print(f\"Fitting: {scale}, {dataset}\")\n",
    "\n",
    "    # data associated to scale and dataset\n",
    "    X, Y = g['X'].to_numpy(), g['Y'].to_numpy()\n",
    "\n",
    "    # fit\n",
    "    res_fit = lg.fit(X, Y, method=\"diff_evol\")\n",
    "    \n",
    "    # calc nllf and beta for fitted pars\n",
    "    res_nllf = lg.nllf(X, Y, res_fit['pars'], jac = True)\n",
    "    beta = lg.get_beta(res_fit['pars'])\n",
    "    \n",
    "    # covariance matrix of pars\n",
    "    cov = lg.cov(X, Y, res_fit['pars'])\n",
    "    \n",
    "    # standard error of pars\n",
    "    pars_SE = lg.get_SE_pars_normal(cov)\n",
    "\n",
    "    # CI of params (assuming asymptotic distr of parameters) : Wald approximation\n",
    "    pars_CI = lg.get_pars_quantiles_normal(probs, res_fit['pars'], cov)\n",
    "    \n",
    "    # goodness of fit\n",
    "    res_gof = lg.goodness_of_fit(X, Y, res_fit['pars'])\n",
    "\n",
    "    lst.append({\"scale\": scale, \"dataset\": dataset, \"beta\": beta} | \n",
    "               {\"nllf\": res_nllf[0], \"jac_norm\": np.linalg.norm(res_nllf[1]), \"jac\": res_nllf[1]} | \n",
    "               {\"cov\": cov, \"cov_evals\": np.linalg.eigvals(cov)} |\n",
    "               {\"SE\": pars_SE, \"LCL\": pars_CI[0], \"UCL\": pars_CI[1]} |\n",
    "               res_gof |res_fit)\n",
    " \n",
    "df_fit = pd.DataFrame(lst)\n",
    "df_fit"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "9e51e07c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['scale', 'dataset', 'beta', 'nllf', 'jac_norm', 'jac', 'cov',\n",
       "       'cov_evals', 'SE', 'LCL', 'UCL', 'LLF', 'AIC', 'BIC', 'A', 'chi2',\n",
       "       'p-value(chi2)', 'n', 'k', 'dof', 'pars', 'cost', 'success'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_fit.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "1a0a518b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.microsoft.datawrangler.viewer.v0+json": {
       "columns": [
        {
         "name": "index",
         "rawType": "int64",
         "type": "integer"
        },
        {
         "name": "scale",
         "rawType": "object",
         "type": "string"
        },
        {
         "name": "dataset",
         "rawType": "object",
         "type": "string"
        },
        {
         "name": "AIC",
         "rawType": "float64",
         "type": "float"
        },
        {
         "name": "BIC",
         "rawType": "float64",
         "type": "float"
        },
        {
         "name": "A",
         "rawType": "float64",
         "type": "float"
        },
        {
         "name": "n",
         "rawType": "int64",
         "type": "integer"
        },
        {
         "name": "k",
         "rawType": "int64",
         "type": "integer"
        },
        {
         "name": "jac_norm",
         "rawType": "float64",
         "type": "float"
        },
        {
         "name": "dof",
         "rawType": "int64",
         "type": "integer"
        },
        {
         "name": "cost",
         "rawType": "float64",
         "type": "float"
        },
        {
         "name": "chi2",
         "rawType": "float64",
         "type": "float"
        },
        {
         "name": "p-value(chi2)",
         "rawType": "float64",
         "type": "float"
        },
        {
         "name": "success",
         "rawType": "bool",
         "type": "boolean"
        }
       ],
       "ref": "fd36fabd-2706-432f-9c2d-40ea0d86c1b4",
       "rows": [
        [
         "0",
         "log",
         "FULL",
         "17.205940435628484",
         "25.44771247781416",
         "0.9482758620689655",
         "58",
         "4",
         "9.027086951330198e-05",
         "54",
         "4.605007497171726",
         "8.316716016778969",
         "0.9999999999999134",
         "True"
        ],
        [
         "1",
         "log",
         "TRIM",
         "11.49204591378204",
         "19.66425098512024",
         "0.9824561403508771",
         "57",
         "4",
         "0.0026576468930959263",
         "53",
         "1.7901864307580302",
         "2.5851621524449944",
         "1.0",
         "True"
        ],
        [
         "2",
         "plain",
         "FULL",
         "17.53300199015373",
         "25.7747740323394",
         "0.9655172413793104",
         "58",
         "4",
         "0.00024992369877456034",
         "54",
         "4.782069565923425",
         "9.164952239066789",
         "0.9999999999992063",
         "True"
        ],
        [
         "3",
         "plain",
         "TRIM",
         "13.04206802909863",
         "21.21427310043683",
         "0.9649122807017544",
         "57",
         "4",
         "0.0010641844799308764",
         "53",
         "2.5614157053734736",
         "4.605672178037899",
         "1.0",
         "True"
        ]
       ],
       "shape": {
        "columns": 13,
        "rows": 4
       }
      },
      "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>scale</th>\n",
       "      <th>dataset</th>\n",
       "      <th>AIC</th>\n",
       "      <th>BIC</th>\n",
       "      <th>A</th>\n",
       "      <th>n</th>\n",
       "      <th>k</th>\n",
       "      <th>jac_norm</th>\n",
       "      <th>dof</th>\n",
       "      <th>cost</th>\n",
       "      <th>chi2</th>\n",
       "      <th>p-value(chi2)</th>\n",
       "      <th>success</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>log</td>\n",
       "      <td>FULL</td>\n",
       "      <td>17.205940</td>\n",
       "      <td>25.447712</td>\n",
       "      <td>0.948276</td>\n",
       "      <td>58</td>\n",
       "      <td>4</td>\n",
       "      <td>0.000090</td>\n",
       "      <td>54</td>\n",
       "      <td>4.605007</td>\n",
       "      <td>8.316716</td>\n",
       "      <td>1.0</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>log</td>\n",
       "      <td>TRIM</td>\n",
       "      <td>11.492046</td>\n",
       "      <td>19.664251</td>\n",
       "      <td>0.982456</td>\n",
       "      <td>57</td>\n",
       "      <td>4</td>\n",
       "      <td>0.002658</td>\n",
       "      <td>53</td>\n",
       "      <td>1.790186</td>\n",
       "      <td>2.585162</td>\n",
       "      <td>1.0</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>plain</td>\n",
       "      <td>FULL</td>\n",
       "      <td>17.533002</td>\n",
       "      <td>25.774774</td>\n",
       "      <td>0.965517</td>\n",
       "      <td>58</td>\n",
       "      <td>4</td>\n",
       "      <td>0.000250</td>\n",
       "      <td>54</td>\n",
       "      <td>4.782070</td>\n",
       "      <td>9.164952</td>\n",
       "      <td>1.0</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>plain</td>\n",
       "      <td>TRIM</td>\n",
       "      <td>13.042068</td>\n",
       "      <td>21.214273</td>\n",
       "      <td>0.964912</td>\n",
       "      <td>57</td>\n",
       "      <td>4</td>\n",
       "      <td>0.001064</td>\n",
       "      <td>53</td>\n",
       "      <td>2.561416</td>\n",
       "      <td>4.605672</td>\n",
       "      <td>1.0</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   scale dataset        AIC        BIC         A   n  k  jac_norm  dof  \\\n",
       "0    log    FULL  17.205940  25.447712  0.948276  58  4  0.000090   54   \n",
       "1    log    TRIM  11.492046  19.664251  0.982456  57  4  0.002658   53   \n",
       "2  plain    FULL  17.533002  25.774774  0.965517  58  4  0.000250   54   \n",
       "3  plain    TRIM  13.042068  21.214273  0.964912  57  4  0.001064   53   \n",
       "\n",
       "       cost      chi2  p-value(chi2)  success  \n",
       "0  4.605007  8.316716            1.0     True  \n",
       "1  1.790186  2.585162            1.0     True  \n",
       "2  4.782070  9.164952            1.0     True  \n",
       "3  2.561416  4.605672            1.0     True  "
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# goodness of fit measures\n",
    "df_fit[['scale', 'dataset', 'AIC', 'BIC', 'A', 'n', 'k', 'jac_norm', 'dof', 'cost','chi2', \"p-value(chi2)\", 'success']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "96acc3ab",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.microsoft.datawrangler.viewer.v0+json": {
       "columns": [
        {
         "name": "index",
         "rawType": "int64",
         "type": "integer"
        },
        {
         "name": "scale",
         "rawType": "object",
         "type": "string"
        },
        {
         "name": "dataset",
         "rawType": "object",
         "type": "string"
        },
        {
         "name": "idx",
         "rawType": "object",
         "type": "unknown"
        },
        {
         "name": "beta",
         "rawType": "object",
         "type": "unknown"
        },
        {
         "name": "pars",
         "rawType": "object",
         "type": "unknown"
        },
        {
         "name": "SE",
         "rawType": "object",
         "type": "unknown"
        },
        {
         "name": "LCL",
         "rawType": "object",
         "type": "unknown"
        },
        {
         "name": "UCL",
         "rawType": "object",
         "type": "unknown"
        }
       ],
       "ref": "8454420d-2aea-4297-a63f-e280202a0544",
       "rows": [
        [
         "0",
         "log",
         "FULL",
         "0",
         "-40.70878972735575",
         "-40.70878972735575",
         "9.158810285950853",
         "-58.65972802905442",
         "-22.757851425657083"
        ],
        [
         "1",
         "log",
         "FULL",
         "1",
         "152.89575359163732",
         "-3.246147934140206e-08",
         "2.01506281856626",
         "-3.949450583437119",
         "3.9494505185141597"
        ],
        [
         "2",
         "log",
         "FULL",
         "2",
         "-186.37209570461724",
         "15.072426573907938",
         "2.088222696696888",
         "10.979585296682927",
         "19.16526785113295"
        ],
        [
         "3",
         "log",
         "FULL",
         "3",
         "75.72601427528205",
         "-12.365102247520532",
         "1.2823198164234106",
         "-14.878402904372432",
         "-9.851801590668634"
        ],
        [
         "4",
         "log",
         "TRIM",
         "0",
         "-200.0",
         "-200.0",
         "59.349028637920135",
         "-316.32195864775974",
         "-83.67804135224027"
        ],
        [
         "5",
         "log",
         "TRIM",
         "1",
         "1039.2599847864008",
         "5.358909598118482e-06",
         "7.051661754923553",
         "-13.820997711899082",
         "13.821008429718276"
        ],
        [
         "6",
         "log",
         "TRIM",
         "2",
         "-1801.9074537027811",
         "-55.89466774410729",
         "8.593492705609279",
         "-72.73760394850915",
         "-39.051731539705436"
        ],
        [
         "7",
         "log",
         "TRIM",
         "3",
         "1041.4046274080492",
         "32.23755550264896",
         "4.562186838423819",
         "23.295833608595615",
         "41.1792773967023"
        ],
        [
         "8",
         "plain",
         "FULL",
         "0",
         "-123.99037983249336",
         "-123.99037983249336",
         "18.872851531659897",
         "-160.98048912011836",
         "-87.00027054486836"
        ],
        [
         "9",
         "plain",
         "FULL",
         "1",
         "163.79100271566466",
         "-8.906667599732129e-07",
         "1.3204975195291813",
         "-2.5881284706184324",
         "2.588126689284912"
        ],
        [
         "10",
         "plain",
         "FULL",
         "2",
         "-71.44674345215762",
         "5.582611650509819",
         "0.4734971796398773",
         "4.654574231634367",
         "6.510649069385271"
        ],
        [
         "11",
         "plain",
         "FULL",
         "3",
         "10.388517613469324",
         "-12.79808590046433",
         "0.888416494563243",
         "-14.539350233079611",
         "-11.056821567849049"
        ],
        [
         "12",
         "plain",
         "TRIM",
         "0",
         "-200.0",
         "-200.0",
         "327.3325220110837",
         "-841.5599541103887",
         "441.5599541103886"
        ],
        [
         "13",
         "plain",
         "TRIM",
         "1",
         "300.1738053663387",
         "-3.143391359711194",
         "1.6616685150655168",
         "-6.40020180348376",
         "0.11341908406137113"
        ],
        [
         "14",
         "plain",
         "TRIM",
         "2",
         "-153.83046339703475",
         "9.02867757719169",
         "10.194270280593756",
         "-10.951725021439106",
         "29.009080175822483"
        ],
        [
         "15",
         "plain",
         "TRIM",
         "3",
         "27.172339597628",
         "-17.03798392199123",
         "16.60650632409131",
         "-49.586138226246845",
         "15.51017038226438"
        ]
       ],
       "shape": {
        "columns": 8,
        "rows": 16
       }
      },
      "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>scale</th>\n",
       "      <th>dataset</th>\n",
       "      <th>idx</th>\n",
       "      <th>beta</th>\n",
       "      <th>pars</th>\n",
       "      <th>SE</th>\n",
       "      <th>LCL</th>\n",
       "      <th>UCL</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>log</td>\n",
       "      <td>FULL</td>\n",
       "      <td>0</td>\n",
       "      <td>-40.70879</td>\n",
       "      <td>-40.70879</td>\n",
       "      <td>9.15881</td>\n",
       "      <td>-58.659728</td>\n",
       "      <td>-22.757851</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>log</td>\n",
       "      <td>FULL</td>\n",
       "      <td>1</td>\n",
       "      <td>152.895754</td>\n",
       "      <td>-0.0</td>\n",
       "      <td>2.015063</td>\n",
       "      <td>-3.949451</td>\n",
       "      <td>3.949451</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>log</td>\n",
       "      <td>FULL</td>\n",
       "      <td>2</td>\n",
       "      <td>-186.372096</td>\n",
       "      <td>15.072427</td>\n",
       "      <td>2.088223</td>\n",
       "      <td>10.979585</td>\n",
       "      <td>19.165268</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>log</td>\n",
       "      <td>FULL</td>\n",
       "      <td>3</td>\n",
       "      <td>75.726014</td>\n",
       "      <td>-12.365102</td>\n",
       "      <td>1.28232</td>\n",
       "      <td>-14.878403</td>\n",
       "      <td>-9.851802</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>log</td>\n",
       "      <td>TRIM</td>\n",
       "      <td>0</td>\n",
       "      <td>-200.0</td>\n",
       "      <td>-200.0</td>\n",
       "      <td>59.349029</td>\n",
       "      <td>-316.321959</td>\n",
       "      <td>-83.678041</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>log</td>\n",
       "      <td>TRIM</td>\n",
       "      <td>1</td>\n",
       "      <td>1039.259985</td>\n",
       "      <td>0.000005</td>\n",
       "      <td>7.051662</td>\n",
       "      <td>-13.820998</td>\n",
       "      <td>13.821008</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>log</td>\n",
       "      <td>TRIM</td>\n",
       "      <td>2</td>\n",
       "      <td>-1801.907454</td>\n",
       "      <td>-55.894668</td>\n",
       "      <td>8.593493</td>\n",
       "      <td>-72.737604</td>\n",
       "      <td>-39.051732</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>log</td>\n",
       "      <td>TRIM</td>\n",
       "      <td>3</td>\n",
       "      <td>1041.404627</td>\n",
       "      <td>32.237556</td>\n",
       "      <td>4.562187</td>\n",
       "      <td>23.295834</td>\n",
       "      <td>41.179277</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>plain</td>\n",
       "      <td>FULL</td>\n",
       "      <td>0</td>\n",
       "      <td>-123.99038</td>\n",
       "      <td>-123.99038</td>\n",
       "      <td>18.872852</td>\n",
       "      <td>-160.980489</td>\n",
       "      <td>-87.000271</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>plain</td>\n",
       "      <td>FULL</td>\n",
       "      <td>1</td>\n",
       "      <td>163.791003</td>\n",
       "      <td>-0.000001</td>\n",
       "      <td>1.320498</td>\n",
       "      <td>-2.588128</td>\n",
       "      <td>2.588127</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>plain</td>\n",
       "      <td>FULL</td>\n",
       "      <td>2</td>\n",
       "      <td>-71.446743</td>\n",
       "      <td>5.582612</td>\n",
       "      <td>0.473497</td>\n",
       "      <td>4.654574</td>\n",
       "      <td>6.510649</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>plain</td>\n",
       "      <td>FULL</td>\n",
       "      <td>3</td>\n",
       "      <td>10.388518</td>\n",
       "      <td>-12.798086</td>\n",
       "      <td>0.888416</td>\n",
       "      <td>-14.53935</td>\n",
       "      <td>-11.056822</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>plain</td>\n",
       "      <td>TRIM</td>\n",
       "      <td>0</td>\n",
       "      <td>-200.0</td>\n",
       "      <td>-200.0</td>\n",
       "      <td>327.332522</td>\n",
       "      <td>-841.559954</td>\n",
       "      <td>441.559954</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>plain</td>\n",
       "      <td>TRIM</td>\n",
       "      <td>1</td>\n",
       "      <td>300.173805</td>\n",
       "      <td>-3.143391</td>\n",
       "      <td>1.661669</td>\n",
       "      <td>-6.400202</td>\n",
       "      <td>0.113419</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>plain</td>\n",
       "      <td>TRIM</td>\n",
       "      <td>2</td>\n",
       "      <td>-153.830463</td>\n",
       "      <td>9.028678</td>\n",
       "      <td>10.19427</td>\n",
       "      <td>-10.951725</td>\n",
       "      <td>29.00908</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>plain</td>\n",
       "      <td>TRIM</td>\n",
       "      <td>3</td>\n",
       "      <td>27.17234</td>\n",
       "      <td>-17.037984</td>\n",
       "      <td>16.606506</td>\n",
       "      <td>-49.586138</td>\n",
       "      <td>15.51017</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    scale dataset idx         beta       pars          SE         LCL  \\\n",
       "0     log    FULL   0    -40.70879  -40.70879     9.15881  -58.659728   \n",
       "1     log    FULL   1   152.895754       -0.0    2.015063   -3.949451   \n",
       "2     log    FULL   2  -186.372096  15.072427    2.088223   10.979585   \n",
       "3     log    FULL   3    75.726014 -12.365102     1.28232  -14.878403   \n",
       "4     log    TRIM   0       -200.0     -200.0   59.349029 -316.321959   \n",
       "5     log    TRIM   1  1039.259985   0.000005    7.051662  -13.820998   \n",
       "6     log    TRIM   2 -1801.907454 -55.894668    8.593493  -72.737604   \n",
       "7     log    TRIM   3  1041.404627  32.237556    4.562187   23.295834   \n",
       "8   plain    FULL   0   -123.99038 -123.99038   18.872852 -160.980489   \n",
       "9   plain    FULL   1   163.791003  -0.000001    1.320498   -2.588128   \n",
       "10  plain    FULL   2   -71.446743   5.582612    0.473497    4.654574   \n",
       "11  plain    FULL   3    10.388518 -12.798086    0.888416   -14.53935   \n",
       "12  plain    TRIM   0       -200.0     -200.0  327.332522 -841.559954   \n",
       "13  plain    TRIM   1   300.173805  -3.143391    1.661669   -6.400202   \n",
       "14  plain    TRIM   2  -153.830463   9.028678    10.19427  -10.951725   \n",
       "15  plain    TRIM   3     27.17234 -17.037984   16.606506  -49.586138   \n",
       "\n",
       "           UCL  \n",
       "0   -22.757851  \n",
       "1     3.949451  \n",
       "2    19.165268  \n",
       "3    -9.851802  \n",
       "4   -83.678041  \n",
       "5    13.821008  \n",
       "6   -39.051732  \n",
       "7    41.179277  \n",
       "8   -87.000271  \n",
       "9     2.588127  \n",
       "10    6.510649  \n",
       "11  -11.056822  \n",
       "12  441.559954  \n",
       "13    0.113419  \n",
       "14    29.00908  \n",
       "15    15.51017  "
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# table of parameters and their CI (Wald approximation)\n",
    "df_pars = df_fit[[\"scale\", \"dataset\", \"beta\", \"pars\", \"SE\", \"LCL\", \"UCL\"]].copy()\n",
    "\n",
    "# add index column for exploding\n",
    "indices = df_pars[\"pars\"].apply(lambda x: list(range(len(x))))\n",
    "df_pars.insert(2, \"idx\", indices)\n",
    "\n",
    "# convert to long format by exploding the pars, SE, LCL, UCL, beta columns\n",
    "cols = [\"idx\", \"beta\", \"pars\", \"SE\", \"LCL\", \"UCL\"]\n",
    "df_pars_long = df_pars.explode(cols, ignore_index=True)\n",
    "df_pars_long"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "7438f05d",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Plotting: log, FULL\n",
      "Plotting: log, TRIM\n",
      "Plotting: plain, FULL\n",
      "Plotting: plain, TRIM\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x500 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# x axis labels for plots\n",
    "xlabs = {\"plain\": r\"$X$\", \"log\": r\"$\\log(X)$\"}\n",
    "\n",
    "# set index for easier access to fit parameters\n",
    "df_fit_index = df_fit.set_index([\"scale\", \"dataset\"])\n",
    "\n",
    "plot_labs = dict(zip(scales[::-1], ['A', 'B']))\n",
    "\n",
    "#\n",
    "# plotting data and fits\n",
    "#\n",
    "\n",
    "fig, axs = plt.subplots(ncols = len(scales), figsize = (6*len(scales), 5))\n",
    "\n",
    "for (scale, dataset), g in df_data.groupby(['scale', 'dataset']):\n",
    "    \n",
    "    print(f\"Plotting: {scale}, {dataset}\")\n",
    "\n",
    "    # data associated to scale and dataset\n",
    "    X, Y = g['X'].to_numpy(), g['Y'].to_numpy()\n",
    "\n",
    "    # choose axis and set labels\n",
    "    ax = axs[0 if scale == \"log\" else 1]\n",
    "    ax.set_xlabel(xlabs[scale])\n",
    "    ax.set_ylabel(\"P(AE|X = x)\")\n",
    "\n",
    "    # label on plots\n",
    "    ax.text(0.05, 0.9, f\"{plot_labs[scale]}\", transform=ax.transAxes)\n",
    "\n",
    "    # data\n",
    "    for i, lab in enumerate([\"NC\", \"AE\"]): \n",
    "        ax.scatter(X[Y == i], Y[Y == i], label = f'data: {dataset} {lab}', alpha = 0.5)\n",
    "\n",
    "    # fitted curve\n",
    "    fit_pars = df_fit_index.loc[(scale, dataset), \"pars\"]\n",
    "    x_fit = np.linspace(X.min(), X.max(), 100)\n",
    "    y_fit = lg.model(x_fit, fit_pars)\n",
    "\n",
    "    ax.plot(x_fit, y_fit, label = f\"fit {dataset}\")\n",
    " \n",
    "handles, labels = ax.get_legend_handles_labels()\n",
    "fig.legend(handles, labels, bbox_to_anchor=(1.05, 0.5), loc='center right')\n",
    "\n",
    "plt.savefig(os.path.join(results_path, \"logit_fit_paper.pdf\"))\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "bda6a88d",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{('log', 'FULL'): 'A', ('plain', 'FULL'): 'B', ('log', 'TRIM'): 'C', ('plain', 'TRIM'): 'D'}\n",
      "Plotting: log, FULL\n",
      "Plotting: log, TRIM\n",
      "Plotting: plain, FULL\n",
      "Plotting: plain, TRIM\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x1000 with 4 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#\n",
    "# plotting data and fits + simple confidence intervals\n",
    "#\n",
    "\n",
    "fig, axs = plt.subplots(ncols = len(scales), nrows = len(datasets), \n",
    "                        figsize = (6*len(scales), 5*len(datasets)))\n",
    "\n",
    "plot_labs = dict(zip([(scale, dataset) for dataset in datasets for scale in scales[::-1]], ['A','B','C','D']))\n",
    "print(plot_labs)\n",
    "\n",
    "for (scale, dataset), g in df_data.groupby(['scale', 'dataset']):\n",
    "    \n",
    "    print(f\"Plotting: {scale}, {dataset}\")\n",
    "\n",
    "    # data associated to scale and dataset\n",
    "    X, Y = g['X'].to_numpy(), g['Y'].to_numpy()\n",
    "\n",
    "    # choose axis and set labels\n",
    "    ax = axs[0 if dataset == \"FULL\" else 1, 0 if scale == \"log\" else 1]\n",
    "    ax.set_xlabel(xlabs[scale])\n",
    "    ax.set_ylabel(\"P(AE|X = x)\")\n",
    "\n",
    "    # label on plots    \n",
    "    ax.text(0.05, 0.9, f\"{plot_labs[(scale, dataset)]}\", transform=ax.transAxes)\n",
    "\n",
    "    # data\n",
    "    for i, lab in enumerate([\"NC\", \"AE\"]): \n",
    "        ax.scatter(X[Y == i], Y[Y == i], label = f'data: {lab}', alpha = 0.5)\n",
    "    \n",
    "    # get fit parameters and covariance matrix for scale and dataset\n",
    "    fit_pars = df_fit_index.loc[(scale, dataset), \"pars\"]\n",
    "    fit_cov = df_fit_index.loc[(scale, dataset), \"cov\"]\n",
    "    \n",
    "    # plotting fitted curve\n",
    "    x_fit = np.linspace(X.min(), X.max(), 100)\n",
    "    y_fit = lg.model(x_fit, fit_pars)\n",
    "    ax.plot(x_fit, y_fit, label = f\"fit\")\n",
    "\n",
    "    # plotting simple confidence intervals of quantiles of the model at probs\n",
    "    for method, c in zip([\"normal\", \"delta\"], [\"red\", \"green\"]):\n",
    "        \n",
    "        # define quantile model\n",
    "        fname = f\"lg.get_model_quantiles_{method}\"\n",
    "        \n",
    "        # get result of model quantiles at probs\n",
    "        res = eval(fname)(x_fit, probs, fit_pars, fit_cov)\n",
    "\n",
    "        # make plot of quantiles\n",
    "        ax.fill_between(x_fit, *res, alpha = 0.5, color = c, label = f\"CI: {method}\")\n",
    " \n",
    "handles, labels = ax.get_legend_handles_labels()\n",
    "fig.legend(handles, labels, bbox_to_anchor=(1.05, 0.5), loc='center right')\n",
    "\n",
    "plt.savefig(os.path.join(results_path, \"logit_fit_CI_simple_paper.pdf\"))\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "67fa2c6c",
   "metadata": {},
   "source": [
    "##  Model CI"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "00b2819a",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{('log', 'FULL'): 'A', ('plain', 'FULL'): 'B', ('log', 'TRIM'): 'C', ('plain', 'TRIM'): 'D'}\n",
      "Plotting: log, FULL\n",
      "\tmodel CI:scale:log, dataset:FULL, method:normal\n",
      "\tmodel CI:scale:log, dataset:FULL, method:nonparam_boots\n",
      "\tmodel CI:scale:log, dataset:FULL, method:nonparam_stratified_boots\n",
      "\tmodel CI:scale:log, dataset:FULL, method:parametric_boots\n",
      "Plotting: log, TRIM\n",
      "\tmodel CI:scale:log, dataset:TRIM, method:normal\n",
      "\tmodel CI:scale:log, dataset:TRIM, method:nonparam_boots\n",
      "\tmodel CI:scale:log, dataset:TRIM, method:nonparam_stratified_boots\n",
      "\tmodel CI:scale:log, dataset:TRIM, method:parametric_boots\n",
      "Plotting: plain, FULL\n",
      "\tmodel CI:scale:plain, dataset:FULL, method:normal\n",
      "\tmodel CI:scale:plain, dataset:FULL, method:nonparam_boots\n",
      "\tmodel CI:scale:plain, dataset:FULL, method:nonparam_stratified_boots\n",
      "\tmodel CI:scale:plain, dataset:FULL, method:parametric_boots\n",
      "Plotting: plain, TRIM\n",
      "\tmodel CI:scale:plain, dataset:TRIM, method:normal\n",
      "\tmodel CI:scale:plain, dataset:TRIM, method:nonparam_boots\n",
      "\tmodel CI:scale:plain, dataset:TRIM, method:nonparam_stratified_boots\n",
      "\tmodel CI:scale:plain, dataset:TRIM, method:parametric_boots\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x1000 with 4 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#\n",
    "# plotting data and fits + full set of confidence intervals\n",
    "#\n",
    "\n",
    "fig, axs = plt.subplots(ncols = len(scales), nrows = len(datasets), \n",
    "                        figsize = (6*len(scales), 5*len(datasets)))\n",
    "\n",
    "plot_labs = dict(zip([(scale, dataset) for dataset in datasets for scale in scales[::-1]], ['A','B','C','D']))\n",
    "print(plot_labs)\n",
    "\n",
    "n_boots = 1000\n",
    "boots_pars_results = dict()\n",
    "\n",
    "for (scale, dataset), g in df_data.groupby(['scale', 'dataset']):\n",
    "    \n",
    "    print(f\"Plotting: {scale}, {dataset}\")\n",
    "\n",
    "    # data associated to scale and dataset\n",
    "    X, Y = g['X'].to_numpy(), g['Y'].to_numpy()\n",
    "\n",
    "    # choose axis and set labels\n",
    "    ax = axs[0 if dataset == \"FULL\" else 1, 0 if scale == \"log\" else 1]\n",
    "    ax.set_xlabel(xlabs[scale])\n",
    "    ax.set_ylabel(\"P(AE|X = x)\")\n",
    "    \n",
    "    #ax.text(0.05, 0.9, f\"{plot_labs[(scale, dataset)]}:{scale}, {dataset}\", transform=ax.transAxes)\n",
    "    ax.text(0.05, 0.9, f\"{plot_labs[(scale, dataset)]}\", transform=ax.transAxes)\n",
    "\n",
    "    # data\n",
    "    for i, lab in enumerate([\"NC\", \"AE\"]): \n",
    "        ax.scatter(X[Y == i], Y[Y == i], label = f'data: {lab}', alpha = 0.5)\n",
    "    \n",
    "    # get fit parameters and covariance matrix for scale and dataset\n",
    "    fit_pars = df_fit_index.loc[(scale, dataset), \"pars\"]\n",
    "    fit_cov = df_fit_index.loc[(scale, dataset), \"cov\"]\n",
    "    \n",
    "    # plotting fitted curve\n",
    "    x_fit = np.linspace(X.min(), X.max(), 100)\n",
    "    y_fit = lg.model(x_fit, fit_pars)\n",
    "    ax.plot(x_fit, y_fit, label = f\"fit\")\n",
    "\n",
    "    # plotting simple confidence intervals of quantiles of the model at probs\n",
    "    for method, c in zip([\"normal\", \"delta\"], [\"red\", \"green\"]):\n",
    "        \n",
    "        # define quantile model\n",
    "        fname = f\"lg.get_model_quantiles_{method}\"\n",
    "        \n",
    "        # get result of model quantiles at probs\n",
    "        res = eval(fname)(x_fit, probs, fit_pars, fit_cov)\n",
    "\n",
    "        # make plot of quantiles\n",
    "        ax.fill_between(x_fit, *res, alpha = 0.5, label = f\"CI:{lab}\")\n",
    "\n",
    "    # plot quantiles of models at bootstapped parameters\n",
    "    for method, c in zip(\n",
    "        [\"normal\", \"nonparam_boots\", \"nonparam_stratified_boots\", \"parametric_boots\"],\n",
    "        [\"orange\", \"pink\", \"purple\", \"cyan\"]):\n",
    "    \n",
    "        print(f\"\\tmodel CI: scale:{scale}, dataset:{dataset}, method:{method}\")\n",
    "\n",
    "        # generate boostrapped parameters\n",
    "        fname = f\"lg.get_{method}_pars\"\n",
    "        bpars = eval(fname)(X, Y, m = n_boots)\n",
    "        boots_pars_results[(scale, dataset, method)] = bpars\n",
    "\n",
    "        # quantiles\n",
    "        quant = np.quantile([lg.model(x_fit, p) for p in bpars], probs, axis = 0)\n",
    "\n",
    "        # make plot of quantiles\n",
    "        ax.fill_between(x_fit, *quant, color = c, alpha = 0.5, label = f\"CI: boots, {method}\")\n",
    "\n",
    "handles, labels = ax.get_legend_handles_labels()\n",
    "fig.legend(handles, labels, bbox_to_anchor=(1.15, 0.5), loc='center right')\n",
    "\n",
    "plt.savefig(os.path.join(results_path, \"logit_fit_CI_paper.pdf\"))\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5ce16edf",
   "metadata": {},
   "source": [
    "## Parameter CI"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "91a199e8",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['scale', 'dataset', 'idx', 'pars', 'SE', 'LCL', 'UCL', 'beta'], dtype='object')"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_pars.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "id": "18dce684",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Data: log, FULL\n",
      "\tmodel CI: scale:log, dataset:FULL, method:normal\n",
      "\tmodel CI: scale:log, dataset:FULL, method:nonparam_boots\n",
      "\tmodel CI: scale:log, dataset:FULL, method:nonparam_stratified_boots\n",
      "\tmodel CI: scale:log, dataset:FULL, method:parametric_boots\n",
      "Data: log, TRIM\n",
      "\tmodel CI: scale:log, dataset:TRIM, method:normal\n",
      "\tmodel CI: scale:log, dataset:TRIM, method:nonparam_boots\n",
      "\tmodel CI: scale:log, dataset:TRIM, method:nonparam_stratified_boots\n",
      "\tmodel CI: scale:log, dataset:TRIM, method:parametric_boots\n",
      "Data: plain, FULL\n",
      "\tmodel CI: scale:plain, dataset:FULL, method:normal\n",
      "\tmodel CI: scale:plain, dataset:FULL, method:nonparam_boots\n",
      "\tmodel CI: scale:plain, dataset:FULL, method:nonparam_stratified_boots\n",
      "\tmodel CI: scale:plain, dataset:FULL, method:parametric_boots\n",
      "Data: plain, TRIM\n",
      "\tmodel CI: scale:plain, dataset:TRIM, method:normal\n",
      "\tmodel CI: scale:plain, dataset:TRIM, method:nonparam_boots\n",
      "\tmodel CI: scale:plain, dataset:TRIM, method:nonparam_stratified_boots\n",
      "\tmodel CI: scale:plain, dataset:TRIM, method:parametric_boots\n",
      "pars CI: scale:['plain', 'log'], dateset:['FULL', 'TRIM'], method:Wald/delta\n"
     ]
    },
    {
     "data": {
      "application/vnd.microsoft.datawrangler.viewer.v0+json": {
       "columns": [
        {
         "name": "index",
         "rawType": "int64",
         "type": "integer"
        },
        {
         "name": "scale",
         "rawType": "object",
         "type": "string"
        },
        {
         "name": "dataset",
         "rawType": "object",
         "type": "string"
        },
        {
         "name": "idx",
         "rawType": "object",
         "type": "unknown"
        },
        {
         "name": "beta",
         "rawType": "object",
         "type": "unknown"
        },
        {
         "name": "pars",
         "rawType": "object",
         "type": "unknown"
        },
        {
         "name": "SE",
         "rawType": "object",
         "type": "unknown"
        },
        {
         "name": "LCL[Wald]",
         "rawType": "object",
         "type": "unknown"
        },
        {
         "name": "UCL[Wald]",
         "rawType": "object",
         "type": "unknown"
        },
        {
         "name": "LCL[normal]",
         "rawType": "object",
         "type": "unknown"
        },
        {
         "name": "UCL[normal]",
         "rawType": "object",
         "type": "unknown"
        },
        {
         "name": "LCL[nonparam_boots]",
         "rawType": "object",
         "type": "unknown"
        },
        {
         "name": "UCL[nonparam_boots]",
         "rawType": "object",
         "type": "unknown"
        },
        {
         "name": "LCL[nonparam_stratified_boots]",
         "rawType": "object",
         "type": "unknown"
        },
        {
         "name": "UCL[nonparam_stratified_boots]",
         "rawType": "object",
         "type": "unknown"
        },
        {
         "name": "LCL[parametric_boots]",
         "rawType": "object",
         "type": "unknown"
        },
        {
         "name": "UCL[parametric_boots]",
         "rawType": "object",
         "type": "unknown"
        }
       ],
       "ref": "20f5c6f5-9e17-40f2-b497-4fc0a4cc3510",
       "rows": [
        [
         "0",
         "log",
         "FULL",
         "[0, 1, 2, 3]",
         "[ -40.70878972735575  152.89575359163732 -186.37209570461724\n   75.72601427528205]",
         "[-4.0708789727355750e+01 -3.2461479341402061e-08  1.5072426573907938e+01\n -1.2365102247520532e+01]",
         "[9.158810285950853  2.01506281856626   2.088222696696888\n 1.2823198164234106]",
         "[-58.65972802905442   -3.949450583437119  10.979585296682927\n -14.878402904372432]",
         "[-22.757851425657083    3.9494505185141597  19.16526785113295\n  -9.851801590668634 ]",
         "[-57.954811342485705   -4.0138356721103055  10.818640690127243\n -14.836040414760735 ]",
         "[-22.00703211288663    3.952136202358119  19.27827623263433\n  -9.776812451921128]",
         "[-200.                  -5.2513076921704    -55.954182326754676\n  -22.94802910424142 ]",
         "[-6.0731764198611078e+00  7.9006702425172202e-06  2.9050464494882764e+01\n  3.2228856255592511e+01]",
         "[-200.                  -5.586379582626496  -55.345203655452806\n  -22.53526279250084 ]",
         "[-6.0546706543324165e+00  1.0752242965126996e-05  2.8169345609626156e+01\n  3.2150611885910060e+01]",
         "[-200.                  -6.84252684038771   -65.98301332071642\n  -25.709362692652576]",
         "[-6.4691635623654946e+00  4.5760502382159089e-06  3.6338077477231835e+01\n  3.2237552629612246e+01]"
        ],
        [
         "1",
         "log",
         "TRIM",
         "[0, 1, 2, 3]",
         "[ -200.               1039.2599847864008 -1801.9074537027811\n  1041.4046274080492]",
         "[-2.000000000000000e+02  5.358909598118482e-06 -5.589466774410729e+01\n  3.223755550264896e+01]",
         "[59.349028637920135  7.051661754923553  8.593492705609279\n  4.562186838423819]",
         "[-316.32195864775974   -13.820997711899082  -72.73760394850915\n   23.295833608595615]",
         "[-83.67804135224027   13.821008429718276 -39.051731539705436\n  41.1792773967023  ]",
         "[-312.6059065043852    -13.830680784247852  -73.20686005122812\n   22.9476482882322  ]",
         "[-78.44594079766605  14.04644016103211 -38.65815238030465\n  40.87077883343024]",
         "[-200.                 -13.543478132517475  -57.81600261901305\n  -31.80925888529137 ]",
         "[-1.1377163406870030e+01  2.3118936061291926e-05  5.3358981898418449e+01\n  3.2542636967805201e+01]",
         "[-200.                  -8.100436897660323  -56.94407561076863\n  -31.979313837143852]",
         "[-1.1291955344486256e+01  3.9366249367381821e-05  5.4305137427427745e+01\n  3.2429343470656683e+01]",
         "[-1.7135189469391662e+02 -5.4794926705668496e+00 -5.9586185265359141e+01\n -5.3001846252945120e-04]",
         "[-1.0345906923664232e+01  5.9825773953538001e-06  1.1089474031266639e-04\n  2.8444293798300727e+01]"
        ],
        [
         "2",
         "plain",
         "FULL",
         "[0, 1, 2, 3]",
         "[-123.99037983249336   163.79100271566466   -71.44674345215762\n   10.388517613469324]",
         "[-1.2399037983249336e+02 -8.9066675997321292e-07  5.5826116505098193e+00\n -1.2798085900464329e+01]",
         "[18.872851531659897   1.3204975195291813  0.4734971796398773\n  0.888416494563243 ]",
         "[-160.98048912011836     -2.5881284706184324    4.654574231634367\n  -14.539350233079611 ]",
         "[-87.00027054486836    2.588126689284912   6.510649069385271\n -11.056821567849049]",
         "[-159.97720796284332    -2.589885513994309   -6.46924256085864\n   11.016819232143341]",
         "[-84.93732513107615    2.630318410012333  -4.649827563446508\n  14.42039807100364 ]",
         "[-200.                   -3.2744696456486118  -10.949944029961486\n  -16.82965485199096  ]",
         "[-1.1529580405692991e+01  6.4925545216944264e-06  9.0587339680391104e+00\n  1.7473025163589892e+01]",
         "[-200.                   -3.2525041365979135  -10.917825516323896\n  -17.14879609111528  ]",
         "[-1.1662120852624287e+01  4.0844310865007554e-06  9.7332955839031037e+00\n  1.7244572741100139e+01]",
         "[-200.                  -9.281801533246307  -11.884470793332495\n  -16.781905484034596]",
         "[-1.3819940843104332e+01  3.3383531707073148e-05  8.0037859471444630e+00\n  1.9047467629106510e+01]"
        ],
        [
         "3",
         "plain",
         "TRIM",
         "[0, 1, 2, 3]",
         "[-200.                300.1738053663387  -153.83046339703475\n   27.172339597628  ]",
         "[-200.                  -3.143391359711194    9.02867757719169\n  -17.03798392199123 ]",
         "[327.3325220110837      1.6616685150655168  10.194270280593756\n  16.60650632409131  ]",
         "[-841.5599541103887     -6.40020180348376   -10.951725021439106\n  -49.586138226246845]",
         "[4.4155995411038862e+02 1.1341908406137113e-01 2.9009080175822483e+01\n 1.5510170382264381e+01]",
         "[-823.5075854823789     -6.235829299951522  -28.197252051140598\n  -17.72506610289006 ]",
         "[4.7635181232763875e+02 2.0750629232567128e-01 1.2281438943237951e+01\n 4.8242652360618926e+01]",
         "[-200.                   -3.6330043809451076  -11.074746653344315\n  -17.370661998691713 ]",
         "[-2.0578776938003692e+01  1.7454762513942113e-05  1.0755288036488261e+01\n  1.7734121276891397e+01]",
         "[-200.                  -3.647138568942977  -11.070489600104557\n  -17.426095758233693]",
         "[-2.1084448712208758e+01  1.3293032944642905e-05  1.0754764634746433e+01\n  1.7759353154963584e+01]",
         "[-200.                   -3.9340304616089234  -12.071208182878348\n  -19.235026477207434 ]",
         "[-3.0306069506729404e+01  2.4348035458824826e-06  1.2905613920332401e+01\n  1.8703334947364073e+01]"
        ]
       ],
       "shape": {
        "columns": 16,
        "rows": 4
       }
      },
      "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>scale</th>\n",
       "      <th>dataset</th>\n",
       "      <th>idx</th>\n",
       "      <th>beta</th>\n",
       "      <th>pars</th>\n",
       "      <th>SE</th>\n",
       "      <th>LCL[Wald]</th>\n",
       "      <th>UCL[Wald]</th>\n",
       "      <th>LCL[normal]</th>\n",
       "      <th>UCL[normal]</th>\n",
       "      <th>LCL[nonparam_boots]</th>\n",
       "      <th>UCL[nonparam_boots]</th>\n",
       "      <th>LCL[nonparam_stratified_boots]</th>\n",
       "      <th>UCL[nonparam_stratified_boots]</th>\n",
       "      <th>LCL[parametric_boots]</th>\n",
       "      <th>UCL[parametric_boots]</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>log</td>\n",
       "      <td>FULL</td>\n",
       "      <td>[0, 1, 2, 3]</td>\n",
       "      <td>[-40.70878972735575, 152.89575359163732, -186....</td>\n",
       "      <td>[-40.70878972735575, -3.246147934140206e-08, 1...</td>\n",
       "      <td>[9.158810285950853, 2.01506281856626, 2.088222...</td>\n",
       "      <td>[-58.65972802905442, -3.949450583437119, 10.97...</td>\n",
       "      <td>[-22.757851425657083, 3.9494505185141597, 19.1...</td>\n",
       "      <td>[-57.954811342485705, -4.0138356721103055, 10....</td>\n",
       "      <td>[-22.00703211288663, 3.952136202358119, 19.278...</td>\n",
       "      <td>[-200.0, -5.2513076921704, -55.954182326754676...</td>\n",
       "      <td>[-6.073176419861108, 7.90067024251722e-06, 29....</td>\n",
       "      <td>[-200.0, -5.586379582626496, -55.3452036554528...</td>\n",
       "      <td>[-6.0546706543324165, 1.0752242965126996e-05, ...</td>\n",
       "      <td>[-200.0, -6.84252684038771, -65.98301332071642...</td>\n",
       "      <td>[-6.469163562365495, 4.576050238215909e-06, 36...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>log</td>\n",
       "      <td>TRIM</td>\n",
       "      <td>[0, 1, 2, 3]</td>\n",
       "      <td>[-200.0, 1039.2599847864008, -1801.90745370278...</td>\n",
       "      <td>[-200.0, 5.358909598118482e-06, -55.8946677441...</td>\n",
       "      <td>[59.349028637920135, 7.051661754923553, 8.5934...</td>\n",
       "      <td>[-316.32195864775974, -13.820997711899082, -72...</td>\n",
       "      <td>[-83.67804135224027, 13.821008429718276, -39.0...</td>\n",
       "      <td>[-312.6059065043852, -13.830680784247852, -73....</td>\n",
       "      <td>[-78.44594079766605, 14.04644016103211, -38.65...</td>\n",
       "      <td>[-200.0, -13.543478132517475, -57.816002619013...</td>\n",
       "      <td>[-11.37716340687003, 2.3118936061291926e-05, 5...</td>\n",
       "      <td>[-200.0, -8.100436897660323, -56.9440756107686...</td>\n",
       "      <td>[-11.291955344486256, 3.936624936738182e-05, 5...</td>\n",
       "      <td>[-171.35189469391662, -5.47949267056685, -59.5...</td>\n",
       "      <td>[-10.345906923664232, 5.9825773953538e-06, 0.0...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>plain</td>\n",
       "      <td>FULL</td>\n",
       "      <td>[0, 1, 2, 3]</td>\n",
       "      <td>[-123.99037983249336, 163.79100271566466, -71....</td>\n",
       "      <td>[-123.99037983249336, -8.906667599732129e-07, ...</td>\n",
       "      <td>[18.872851531659897, 1.3204975195291813, 0.473...</td>\n",
       "      <td>[-160.98048912011836, -2.5881284706184324, 4.6...</td>\n",
       "      <td>[-87.00027054486836, 2.588126689284912, 6.5106...</td>\n",
       "      <td>[-159.97720796284332, -2.589885513994309, -6.4...</td>\n",
       "      <td>[-84.93732513107615, 2.630318410012333, -4.649...</td>\n",
       "      <td>[-200.0, -3.2744696456486118, -10.949944029961...</td>\n",
       "      <td>[-11.529580405692991, 6.492554521694426e-06, 9...</td>\n",
       "      <td>[-200.0, -3.2525041365979135, -10.917825516323...</td>\n",
       "      <td>[-11.662120852624287, 4.084431086500755e-06, 9...</td>\n",
       "      <td>[-200.0, -9.281801533246307, -11.8844707933324...</td>\n",
       "      <td>[-13.819940843104332, 3.338353170707315e-05, 8...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>plain</td>\n",
       "      <td>TRIM</td>\n",
       "      <td>[0, 1, 2, 3]</td>\n",
       "      <td>[-200.0, 300.1738053663387, -153.8304633970347...</td>\n",
       "      <td>[-200.0, -3.143391359711194, 9.02867757719169,...</td>\n",
       "      <td>[327.3325220110837, 1.6616685150655168, 10.194...</td>\n",
       "      <td>[-841.5599541103887, -6.40020180348376, -10.95...</td>\n",
       "      <td>[441.5599541103886, 0.11341908406137113, 29.00...</td>\n",
       "      <td>[-823.5075854823789, -6.235829299951522, -28.1...</td>\n",
       "      <td>[476.35181232763875, 0.20750629232567128, 12.2...</td>\n",
       "      <td>[-200.0, -3.6330043809451076, -11.074746653344...</td>\n",
       "      <td>[-20.578776938003692, 1.7454762513942113e-05, ...</td>\n",
       "      <td>[-200.0, -3.647138568942977, -11.0704896001045...</td>\n",
       "      <td>[-21.084448712208758, 1.3293032944642905e-05, ...</td>\n",
       "      <td>[-200.0, -3.9340304616089234, -12.071208182878...</td>\n",
       "      <td>[-30.306069506729404, 2.4348035458824826e-06, ...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   scale dataset           idx  \\\n",
       "0    log    FULL  [0, 1, 2, 3]   \n",
       "1    log    TRIM  [0, 1, 2, 3]   \n",
       "2  plain    FULL  [0, 1, 2, 3]   \n",
       "3  plain    TRIM  [0, 1, 2, 3]   \n",
       "\n",
       "                                                beta  \\\n",
       "0  [-40.70878972735575, 152.89575359163732, -186....   \n",
       "1  [-200.0, 1039.2599847864008, -1801.90745370278...   \n",
       "2  [-123.99037983249336, 163.79100271566466, -71....   \n",
       "3  [-200.0, 300.1738053663387, -153.8304633970347...   \n",
       "\n",
       "                                                pars  \\\n",
       "0  [-40.70878972735575, -3.246147934140206e-08, 1...   \n",
       "1  [-200.0, 5.358909598118482e-06, -55.8946677441...   \n",
       "2  [-123.99037983249336, -8.906667599732129e-07, ...   \n",
       "3  [-200.0, -3.143391359711194, 9.02867757719169,...   \n",
       "\n",
       "                                                  SE  \\\n",
       "0  [9.158810285950853, 2.01506281856626, 2.088222...   \n",
       "1  [59.349028637920135, 7.051661754923553, 8.5934...   \n",
       "2  [18.872851531659897, 1.3204975195291813, 0.473...   \n",
       "3  [327.3325220110837, 1.6616685150655168, 10.194...   \n",
       "\n",
       "                                           LCL[Wald]  \\\n",
       "0  [-58.65972802905442, -3.949450583437119, 10.97...   \n",
       "1  [-316.32195864775974, -13.820997711899082, -72...   \n",
       "2  [-160.98048912011836, -2.5881284706184324, 4.6...   \n",
       "3  [-841.5599541103887, -6.40020180348376, -10.95...   \n",
       "\n",
       "                                           UCL[Wald]  \\\n",
       "0  [-22.757851425657083, 3.9494505185141597, 19.1...   \n",
       "1  [-83.67804135224027, 13.821008429718276, -39.0...   \n",
       "2  [-87.00027054486836, 2.588126689284912, 6.5106...   \n",
       "3  [441.5599541103886, 0.11341908406137113, 29.00...   \n",
       "\n",
       "                                         LCL[normal]  \\\n",
       "0  [-57.954811342485705, -4.0138356721103055, 10....   \n",
       "1  [-312.6059065043852, -13.830680784247852, -73....   \n",
       "2  [-159.97720796284332, -2.589885513994309, -6.4...   \n",
       "3  [-823.5075854823789, -6.235829299951522, -28.1...   \n",
       "\n",
       "                                         UCL[normal]  \\\n",
       "0  [-22.00703211288663, 3.952136202358119, 19.278...   \n",
       "1  [-78.44594079766605, 14.04644016103211, -38.65...   \n",
       "2  [-84.93732513107615, 2.630318410012333, -4.649...   \n",
       "3  [476.35181232763875, 0.20750629232567128, 12.2...   \n",
       "\n",
       "                                 LCL[nonparam_boots]  \\\n",
       "0  [-200.0, -5.2513076921704, -55.954182326754676...   \n",
       "1  [-200.0, -13.543478132517475, -57.816002619013...   \n",
       "2  [-200.0, -3.2744696456486118, -10.949944029961...   \n",
       "3  [-200.0, -3.6330043809451076, -11.074746653344...   \n",
       "\n",
       "                                 UCL[nonparam_boots]  \\\n",
       "0  [-6.073176419861108, 7.90067024251722e-06, 29....   \n",
       "1  [-11.37716340687003, 2.3118936061291926e-05, 5...   \n",
       "2  [-11.529580405692991, 6.492554521694426e-06, 9...   \n",
       "3  [-20.578776938003692, 1.7454762513942113e-05, ...   \n",
       "\n",
       "                      LCL[nonparam_stratified_boots]  \\\n",
       "0  [-200.0, -5.586379582626496, -55.3452036554528...   \n",
       "1  [-200.0, -8.100436897660323, -56.9440756107686...   \n",
       "2  [-200.0, -3.2525041365979135, -10.917825516323...   \n",
       "3  [-200.0, -3.647138568942977, -11.0704896001045...   \n",
       "\n",
       "                      UCL[nonparam_stratified_boots]  \\\n",
       "0  [-6.0546706543324165, 1.0752242965126996e-05, ...   \n",
       "1  [-11.291955344486256, 3.936624936738182e-05, 5...   \n",
       "2  [-11.662120852624287, 4.084431086500755e-06, 9...   \n",
       "3  [-21.084448712208758, 1.3293032944642905e-05, ...   \n",
       "\n",
       "                               LCL[parametric_boots]  \\\n",
       "0  [-200.0, -6.84252684038771, -65.98301332071642...   \n",
       "1  [-171.35189469391662, -5.47949267056685, -59.5...   \n",
       "2  [-200.0, -9.281801533246307, -11.8844707933324...   \n",
       "3  [-200.0, -3.9340304616089234, -12.071208182878...   \n",
       "\n",
       "                               UCL[parametric_boots]  \n",
       "0  [-6.469163562365495, 4.576050238215909e-06, 36...  \n",
       "1  [-10.345906923664232, 5.9825773953538e-06, 0.0...  \n",
       "2  [-13.819940843104332, 3.338353170707315e-05, 8...  \n",
       "3  [-30.306069506729404, 2.4348035458824826e-06, ...  "
      ]
     },
     "execution_count": 87,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dt_res = dict()\n",
    "for (scale, dataset), _ in df_data.groupby(['scale', 'dataset']):\n",
    "    \n",
    "    print(f\"Data: {scale}, {dataset}\")\n",
    "\n",
    "    # plot quantiles of models at bootstapped parameters\n",
    "    for method in [\"normal\", \"nonparam_boots\", \"nonparam_stratified_boots\", \"parametric_boots\"]:\n",
    "    \n",
    "        print(f\"\\tmodel CI: scale:{scale}, dataset:{dataset}, method:{method}\")\n",
    "\n",
    "        # get boostrapped parameters\n",
    "        bpars = boots_pars_results[(scale, dataset, method)]\n",
    "    \n",
    "        # quantiles\n",
    "        quant = np.quantile(bpars, probs, axis = 0)\n",
    "        \n",
    "        if (scale, dataset) not in dt_res: dt_res[(scale, dataset)] = dict()\n",
    "        for lab, q in zip([\"LCL\", \"UCL\"], quant):\n",
    "            dt_res[(scale, dataset)][f\"{lab}[{method}]\"] = q\n",
    "\n",
    "\n",
    "# Joining results of bootstrapped pars quantiles with Wald CI of pars\n",
    "print(f\"pars CI: scale:{scales}, dateset:{datasets}, method:Wald/delta\")\n",
    "\n",
    "method = \"Wald\"\n",
    "df_pars_CI_wald = df_pars.rename(columns={\"LCL\": f\"LCL[{method}]\", \"UCL\": f\"UCL[{method}]\"})\n",
    "df_pars_CI_wald = df_pars_CI_wald.set_index([\"scale\", \"dataset\"])\n",
    "\n",
    "df_res_CI_other = pd.DataFrame(dt_res).T\n",
    "df_res_CI_other.index.names = ['scale', 'dataset']\n",
    "\n",
    "df_pars_CI = df_pars_CI_wald.join(df_res_CI_other)\n",
    "df_pars_CI.reset_index(inplace=True)\n",
    "df_pars_CI"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "id": "e76c0167",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.microsoft.datawrangler.viewer.v0+json": {
       "columns": [
        {
         "name": "index",
         "rawType": "int64",
         "type": "integer"
        },
        {
         "name": "scale",
         "rawType": "object",
         "type": "string"
        },
        {
         "name": "dataset",
         "rawType": "object",
         "type": "string"
        },
        {
         "name": "idx",
         "rawType": "object",
         "type": "unknown"
        },
        {
         "name": "beta",
         "rawType": "object",
         "type": "unknown"
        },
        {
         "name": "pars",
         "rawType": "object",
         "type": "unknown"
        },
        {
         "name": "SE",
         "rawType": "object",
         "type": "unknown"
        },
        {
         "name": "LCL[Wald]",
         "rawType": "object",
         "type": "unknown"
        },
        {
         "name": "UCL[Wald]",
         "rawType": "object",
         "type": "unknown"
        },
        {
         "name": "LCL[normal]",
         "rawType": "object",
         "type": "unknown"
        },
        {
         "name": "UCL[normal]",
         "rawType": "object",
         "type": "unknown"
        },
        {
         "name": "LCL[nonparam_boots]",
         "rawType": "object",
         "type": "unknown"
        },
        {
         "name": "UCL[nonparam_boots]",
         "rawType": "object",
         "type": "unknown"
        },
        {
         "name": "LCL[nonparam_stratified_boots]",
         "rawType": "object",
         "type": "unknown"
        },
        {
         "name": "UCL[nonparam_stratified_boots]",
         "rawType": "object",
         "type": "unknown"
        },
        {
         "name": "LCL[parametric_boots]",
         "rawType": "object",
         "type": "unknown"
        },
        {
         "name": "UCL[parametric_boots]",
         "rawType": "object",
         "type": "unknown"
        }
       ],
       "ref": "fd3914f1-3ccb-43d1-bc6f-601a0c9bebb4",
       "rows": [
        [
         "0",
         "log",
         "FULL",
         "0",
         "-40.70878972735575",
         "-40.70878972735575",
         "9.158810285950853",
         "-58.65972802905442",
         "-22.757851425657083",
         "-57.954811342485705",
         "-22.00703211288663",
         "-200.0",
         "-6.073176419861108",
         "-200.0",
         "-6.0546706543324165",
         "-200.0",
         "-6.469163562365495"
        ],
        [
         "1",
         "log",
         "FULL",
         "1",
         "152.89575359163732",
         "-3.246147934140206e-08",
         "2.01506281856626",
         "-3.949450583437119",
         "3.9494505185141597",
         "-4.0138356721103055",
         "3.952136202358119",
         "-5.2513076921704",
         "7.90067024251722e-06",
         "-5.586379582626496",
         "1.0752242965126996e-05",
         "-6.84252684038771",
         "4.576050238215909e-06"
        ],
        [
         "2",
         "log",
         "FULL",
         "2",
         "-186.37209570461724",
         "15.072426573907938",
         "2.088222696696888",
         "10.979585296682927",
         "19.16526785113295",
         "10.818640690127243",
         "19.27827623263433",
         "-55.954182326754676",
         "29.050464494882764",
         "-55.345203655452806",
         "28.169345609626156",
         "-65.98301332071642",
         "36.338077477231835"
        ],
        [
         "3",
         "log",
         "FULL",
         "3",
         "75.72601427528205",
         "-12.365102247520532",
         "1.2823198164234106",
         "-14.878402904372432",
         "-9.851801590668634",
         "-14.836040414760735",
         "-9.776812451921128",
         "-22.94802910424142",
         "32.22885625559251",
         "-22.53526279250084",
         "32.15061188591006",
         "-25.709362692652576",
         "32.237552629612246"
        ],
        [
         "4",
         "log",
         "TRIM",
         "0",
         "-200.0",
         "-200.0",
         "59.349028637920135",
         "-316.32195864775974",
         "-83.67804135224027",
         "-312.6059065043852",
         "-78.44594079766605",
         "-200.0",
         "-11.37716340687003",
         "-200.0",
         "-11.291955344486256",
         "-171.35189469391662",
         "-10.345906923664232"
        ],
        [
         "5",
         "log",
         "TRIM",
         "1",
         "1039.2599847864008",
         "5.358909598118482e-06",
         "7.051661754923553",
         "-13.820997711899082",
         "13.821008429718276",
         "-13.830680784247852",
         "14.04644016103211",
         "-13.543478132517475",
         "2.3118936061291926e-05",
         "-8.100436897660323",
         "3.936624936738182e-05",
         "-5.47949267056685",
         "5.9825773953538e-06"
        ],
        [
         "6",
         "log",
         "TRIM",
         "2",
         "-1801.9074537027811",
         "-55.89466774410729",
         "8.593492705609279",
         "-72.73760394850915",
         "-39.051731539705436",
         "-73.20686005122812",
         "-38.65815238030465",
         "-57.81600261901305",
         "53.35898189841845",
         "-56.94407561076863",
         "54.305137427427745",
         "-59.58618526535914",
         "0.0001108947403126664"
        ],
        [
         "7",
         "log",
         "TRIM",
         "3",
         "1041.4046274080492",
         "32.23755550264896",
         "4.562186838423819",
         "23.295833608595615",
         "41.1792773967023",
         "22.9476482882322",
         "40.87077883343024",
         "-31.80925888529137",
         "32.5426369678052",
         "-31.979313837143852",
         "32.42934347065668",
         "-0.0005300184625294512",
         "28.444293798300727"
        ],
        [
         "8",
         "plain",
         "FULL",
         "0",
         "-123.99037983249336",
         "-123.99037983249336",
         "18.872851531659897",
         "-160.98048912011836",
         "-87.00027054486836",
         "-159.97720796284332",
         "-84.93732513107615",
         "-200.0",
         "-11.529580405692991",
         "-200.0",
         "-11.662120852624287",
         "-200.0",
         "-13.819940843104332"
        ],
        [
         "9",
         "plain",
         "FULL",
         "1",
         "163.79100271566466",
         "-8.906667599732129e-07",
         "1.3204975195291813",
         "-2.5881284706184324",
         "2.588126689284912",
         "-2.589885513994309",
         "2.630318410012333",
         "-3.2744696456486118",
         "6.492554521694426e-06",
         "-3.2525041365979135",
         "4.084431086500755e-06",
         "-9.281801533246307",
         "3.338353170707315e-05"
        ],
        [
         "10",
         "plain",
         "FULL",
         "2",
         "-71.44674345215762",
         "5.582611650509819",
         "0.4734971796398773",
         "4.654574231634367",
         "6.510649069385271",
         "-6.46924256085864",
         "-4.649827563446508",
         "-10.949944029961486",
         "9.05873396803911",
         "-10.917825516323896",
         "9.733295583903104",
         "-11.884470793332495",
         "8.003785947144463"
        ],
        [
         "11",
         "plain",
         "FULL",
         "3",
         "10.388517613469324",
         "-12.79808590046433",
         "0.888416494563243",
         "-14.539350233079611",
         "-11.056821567849049",
         "11.016819232143341",
         "14.42039807100364",
         "-16.82965485199096",
         "17.47302516358989",
         "-17.14879609111528",
         "17.24457274110014",
         "-16.781905484034596",
         "19.04746762910651"
        ],
        [
         "12",
         "plain",
         "TRIM",
         "0",
         "-200.0",
         "-200.0",
         "327.3325220110837",
         "-841.5599541103887",
         "441.5599541103886",
         "-823.5075854823789",
         "476.35181232763875",
         "-200.0",
         "-20.578776938003692",
         "-200.0",
         "-21.084448712208758",
         "-200.0",
         "-30.306069506729404"
        ],
        [
         "13",
         "plain",
         "TRIM",
         "1",
         "300.1738053663387",
         "-3.143391359711194",
         "1.6616685150655168",
         "-6.40020180348376",
         "0.11341908406137113",
         "-6.235829299951522",
         "0.20750629232567128",
         "-3.6330043809451076",
         "1.7454762513942113e-05",
         "-3.647138568942977",
         "1.3293032944642905e-05",
         "-3.9340304616089234",
         "2.4348035458824826e-06"
        ],
        [
         "14",
         "plain",
         "TRIM",
         "2",
         "-153.83046339703475",
         "9.02867757719169",
         "10.194270280593756",
         "-10.951725021439106",
         "29.009080175822483",
         "-28.197252051140598",
         "12.281438943237951",
         "-11.074746653344315",
         "10.755288036488261",
         "-11.070489600104557",
         "10.754764634746433",
         "-12.071208182878348",
         "12.905613920332401"
        ],
        [
         "15",
         "plain",
         "TRIM",
         "3",
         "27.172339597628",
         "-17.03798392199123",
         "16.60650632409131",
         "-49.586138226246845",
         "15.51017038226438",
         "-17.72506610289006",
         "48.242652360618926",
         "-17.370661998691713",
         "17.734121276891397",
         "-17.426095758233693",
         "17.759353154963584",
         "-19.235026477207434",
         "18.703334947364073"
        ]
       ],
       "shape": {
        "columns": 16,
        "rows": 16
       }
      },
      "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>scale</th>\n",
       "      <th>dataset</th>\n",
       "      <th>idx</th>\n",
       "      <th>beta</th>\n",
       "      <th>pars</th>\n",
       "      <th>SE</th>\n",
       "      <th>LCL[Wald]</th>\n",
       "      <th>UCL[Wald]</th>\n",
       "      <th>LCL[normal]</th>\n",
       "      <th>UCL[normal]</th>\n",
       "      <th>LCL[nonparam_boots]</th>\n",
       "      <th>UCL[nonparam_boots]</th>\n",
       "      <th>LCL[nonparam_stratified_boots]</th>\n",
       "      <th>UCL[nonparam_stratified_boots]</th>\n",
       "      <th>LCL[parametric_boots]</th>\n",
       "      <th>UCL[parametric_boots]</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>log</td>\n",
       "      <td>FULL</td>\n",
       "      <td>0</td>\n",
       "      <td>-40.70879</td>\n",
       "      <td>-40.70879</td>\n",
       "      <td>9.15881</td>\n",
       "      <td>-58.659728</td>\n",
       "      <td>-22.757851</td>\n",
       "      <td>-57.954811</td>\n",
       "      <td>-22.007032</td>\n",
       "      <td>-200.0</td>\n",
       "      <td>-6.073176</td>\n",
       "      <td>-200.0</td>\n",
       "      <td>-6.054671</td>\n",
       "      <td>-200.0</td>\n",
       "      <td>-6.469164</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>log</td>\n",
       "      <td>FULL</td>\n",
       "      <td>1</td>\n",
       "      <td>152.895754</td>\n",
       "      <td>-0.0</td>\n",
       "      <td>2.015063</td>\n",
       "      <td>-3.949451</td>\n",
       "      <td>3.949451</td>\n",
       "      <td>-4.013836</td>\n",
       "      <td>3.952136</td>\n",
       "      <td>-5.251308</td>\n",
       "      <td>0.000008</td>\n",
       "      <td>-5.58638</td>\n",
       "      <td>0.000011</td>\n",
       "      <td>-6.842527</td>\n",
       "      <td>0.000005</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>log</td>\n",
       "      <td>FULL</td>\n",
       "      <td>2</td>\n",
       "      <td>-186.372096</td>\n",
       "      <td>15.072427</td>\n",
       "      <td>2.088223</td>\n",
       "      <td>10.979585</td>\n",
       "      <td>19.165268</td>\n",
       "      <td>10.818641</td>\n",
       "      <td>19.278276</td>\n",
       "      <td>-55.954182</td>\n",
       "      <td>29.050464</td>\n",
       "      <td>-55.345204</td>\n",
       "      <td>28.169346</td>\n",
       "      <td>-65.983013</td>\n",
       "      <td>36.338077</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>log</td>\n",
       "      <td>FULL</td>\n",
       "      <td>3</td>\n",
       "      <td>75.726014</td>\n",
       "      <td>-12.365102</td>\n",
       "      <td>1.28232</td>\n",
       "      <td>-14.878403</td>\n",
       "      <td>-9.851802</td>\n",
       "      <td>-14.83604</td>\n",
       "      <td>-9.776812</td>\n",
       "      <td>-22.948029</td>\n",
       "      <td>32.228856</td>\n",
       "      <td>-22.535263</td>\n",
       "      <td>32.150612</td>\n",
       "      <td>-25.709363</td>\n",
       "      <td>32.237553</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>log</td>\n",
       "      <td>TRIM</td>\n",
       "      <td>0</td>\n",
       "      <td>-200.0</td>\n",
       "      <td>-200.0</td>\n",
       "      <td>59.349029</td>\n",
       "      <td>-316.321959</td>\n",
       "      <td>-83.678041</td>\n",
       "      <td>-312.605907</td>\n",
       "      <td>-78.445941</td>\n",
       "      <td>-200.0</td>\n",
       "      <td>-11.377163</td>\n",
       "      <td>-200.0</td>\n",
       "      <td>-11.291955</td>\n",
       "      <td>-171.351895</td>\n",
       "      <td>-10.345907</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>log</td>\n",
       "      <td>TRIM</td>\n",
       "      <td>1</td>\n",
       "      <td>1039.259985</td>\n",
       "      <td>0.000005</td>\n",
       "      <td>7.051662</td>\n",
       "      <td>-13.820998</td>\n",
       "      <td>13.821008</td>\n",
       "      <td>-13.830681</td>\n",
       "      <td>14.04644</td>\n",
       "      <td>-13.543478</td>\n",
       "      <td>0.000023</td>\n",
       "      <td>-8.100437</td>\n",
       "      <td>0.000039</td>\n",
       "      <td>-5.479493</td>\n",
       "      <td>0.000006</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>log</td>\n",
       "      <td>TRIM</td>\n",
       "      <td>2</td>\n",
       "      <td>-1801.907454</td>\n",
       "      <td>-55.894668</td>\n",
       "      <td>8.593493</td>\n",
       "      <td>-72.737604</td>\n",
       "      <td>-39.051732</td>\n",
       "      <td>-73.20686</td>\n",
       "      <td>-38.658152</td>\n",
       "      <td>-57.816003</td>\n",
       "      <td>53.358982</td>\n",
       "      <td>-56.944076</td>\n",
       "      <td>54.305137</td>\n",
       "      <td>-59.586185</td>\n",
       "      <td>0.000111</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>log</td>\n",
       "      <td>TRIM</td>\n",
       "      <td>3</td>\n",
       "      <td>1041.404627</td>\n",
       "      <td>32.237556</td>\n",
       "      <td>4.562187</td>\n",
       "      <td>23.295834</td>\n",
       "      <td>41.179277</td>\n",
       "      <td>22.947648</td>\n",
       "      <td>40.870779</td>\n",
       "      <td>-31.809259</td>\n",
       "      <td>32.542637</td>\n",
       "      <td>-31.979314</td>\n",
       "      <td>32.429343</td>\n",
       "      <td>-0.00053</td>\n",
       "      <td>28.444294</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>plain</td>\n",
       "      <td>FULL</td>\n",
       "      <td>0</td>\n",
       "      <td>-123.99038</td>\n",
       "      <td>-123.99038</td>\n",
       "      <td>18.872852</td>\n",
       "      <td>-160.980489</td>\n",
       "      <td>-87.000271</td>\n",
       "      <td>-159.977208</td>\n",
       "      <td>-84.937325</td>\n",
       "      <td>-200.0</td>\n",
       "      <td>-11.52958</td>\n",
       "      <td>-200.0</td>\n",
       "      <td>-11.662121</td>\n",
       "      <td>-200.0</td>\n",
       "      <td>-13.819941</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>plain</td>\n",
       "      <td>FULL</td>\n",
       "      <td>1</td>\n",
       "      <td>163.791003</td>\n",
       "      <td>-0.000001</td>\n",
       "      <td>1.320498</td>\n",
       "      <td>-2.588128</td>\n",
       "      <td>2.588127</td>\n",
       "      <td>-2.589886</td>\n",
       "      <td>2.630318</td>\n",
       "      <td>-3.27447</td>\n",
       "      <td>0.000006</td>\n",
       "      <td>-3.252504</td>\n",
       "      <td>0.000004</td>\n",
       "      <td>-9.281802</td>\n",
       "      <td>0.000033</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>plain</td>\n",
       "      <td>FULL</td>\n",
       "      <td>2</td>\n",
       "      <td>-71.446743</td>\n",
       "      <td>5.582612</td>\n",
       "      <td>0.473497</td>\n",
       "      <td>4.654574</td>\n",
       "      <td>6.510649</td>\n",
       "      <td>-6.469243</td>\n",
       "      <td>-4.649828</td>\n",
       "      <td>-10.949944</td>\n",
       "      <td>9.058734</td>\n",
       "      <td>-10.917826</td>\n",
       "      <td>9.733296</td>\n",
       "      <td>-11.884471</td>\n",
       "      <td>8.003786</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>plain</td>\n",
       "      <td>FULL</td>\n",
       "      <td>3</td>\n",
       "      <td>10.388518</td>\n",
       "      <td>-12.798086</td>\n",
       "      <td>0.888416</td>\n",
       "      <td>-14.53935</td>\n",
       "      <td>-11.056822</td>\n",
       "      <td>11.016819</td>\n",
       "      <td>14.420398</td>\n",
       "      <td>-16.829655</td>\n",
       "      <td>17.473025</td>\n",
       "      <td>-17.148796</td>\n",
       "      <td>17.244573</td>\n",
       "      <td>-16.781905</td>\n",
       "      <td>19.047468</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>plain</td>\n",
       "      <td>TRIM</td>\n",
       "      <td>0</td>\n",
       "      <td>-200.0</td>\n",
       "      <td>-200.0</td>\n",
       "      <td>327.332522</td>\n",
       "      <td>-841.559954</td>\n",
       "      <td>441.559954</td>\n",
       "      <td>-823.507585</td>\n",
       "      <td>476.351812</td>\n",
       "      <td>-200.0</td>\n",
       "      <td>-20.578777</td>\n",
       "      <td>-200.0</td>\n",
       "      <td>-21.084449</td>\n",
       "      <td>-200.0</td>\n",
       "      <td>-30.30607</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>plain</td>\n",
       "      <td>TRIM</td>\n",
       "      <td>1</td>\n",
       "      <td>300.173805</td>\n",
       "      <td>-3.143391</td>\n",
       "      <td>1.661669</td>\n",
       "      <td>-6.400202</td>\n",
       "      <td>0.113419</td>\n",
       "      <td>-6.235829</td>\n",
       "      <td>0.207506</td>\n",
       "      <td>-3.633004</td>\n",
       "      <td>0.000017</td>\n",
       "      <td>-3.647139</td>\n",
       "      <td>0.000013</td>\n",
       "      <td>-3.93403</td>\n",
       "      <td>0.000002</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>plain</td>\n",
       "      <td>TRIM</td>\n",
       "      <td>2</td>\n",
       "      <td>-153.830463</td>\n",
       "      <td>9.028678</td>\n",
       "      <td>10.19427</td>\n",
       "      <td>-10.951725</td>\n",
       "      <td>29.00908</td>\n",
       "      <td>-28.197252</td>\n",
       "      <td>12.281439</td>\n",
       "      <td>-11.074747</td>\n",
       "      <td>10.755288</td>\n",
       "      <td>-11.07049</td>\n",
       "      <td>10.754765</td>\n",
       "      <td>-12.071208</td>\n",
       "      <td>12.905614</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>plain</td>\n",
       "      <td>TRIM</td>\n",
       "      <td>3</td>\n",
       "      <td>27.17234</td>\n",
       "      <td>-17.037984</td>\n",
       "      <td>16.606506</td>\n",
       "      <td>-49.586138</td>\n",
       "      <td>15.51017</td>\n",
       "      <td>-17.725066</td>\n",
       "      <td>48.242652</td>\n",
       "      <td>-17.370662</td>\n",
       "      <td>17.734121</td>\n",
       "      <td>-17.426096</td>\n",
       "      <td>17.759353</td>\n",
       "      <td>-19.235026</td>\n",
       "      <td>18.703335</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    scale dataset idx         beta       pars          SE   LCL[Wald]  \\\n",
       "0     log    FULL   0    -40.70879  -40.70879     9.15881  -58.659728   \n",
       "1     log    FULL   1   152.895754       -0.0    2.015063   -3.949451   \n",
       "2     log    FULL   2  -186.372096  15.072427    2.088223   10.979585   \n",
       "3     log    FULL   3    75.726014 -12.365102     1.28232  -14.878403   \n",
       "4     log    TRIM   0       -200.0     -200.0   59.349029 -316.321959   \n",
       "5     log    TRIM   1  1039.259985   0.000005    7.051662  -13.820998   \n",
       "6     log    TRIM   2 -1801.907454 -55.894668    8.593493  -72.737604   \n",
       "7     log    TRIM   3  1041.404627  32.237556    4.562187   23.295834   \n",
       "8   plain    FULL   0   -123.99038 -123.99038   18.872852 -160.980489   \n",
       "9   plain    FULL   1   163.791003  -0.000001    1.320498   -2.588128   \n",
       "10  plain    FULL   2   -71.446743   5.582612    0.473497    4.654574   \n",
       "11  plain    FULL   3    10.388518 -12.798086    0.888416   -14.53935   \n",
       "12  plain    TRIM   0       -200.0     -200.0  327.332522 -841.559954   \n",
       "13  plain    TRIM   1   300.173805  -3.143391    1.661669   -6.400202   \n",
       "14  plain    TRIM   2  -153.830463   9.028678    10.19427  -10.951725   \n",
       "15  plain    TRIM   3     27.17234 -17.037984   16.606506  -49.586138   \n",
       "\n",
       "     UCL[Wald] LCL[normal] UCL[normal] LCL[nonparam_boots]  \\\n",
       "0   -22.757851  -57.954811  -22.007032              -200.0   \n",
       "1     3.949451   -4.013836    3.952136           -5.251308   \n",
       "2    19.165268   10.818641   19.278276          -55.954182   \n",
       "3    -9.851802   -14.83604   -9.776812          -22.948029   \n",
       "4   -83.678041 -312.605907  -78.445941              -200.0   \n",
       "5    13.821008  -13.830681    14.04644          -13.543478   \n",
       "6   -39.051732   -73.20686  -38.658152          -57.816003   \n",
       "7    41.179277   22.947648   40.870779          -31.809259   \n",
       "8   -87.000271 -159.977208  -84.937325              -200.0   \n",
       "9     2.588127   -2.589886    2.630318            -3.27447   \n",
       "10    6.510649   -6.469243   -4.649828          -10.949944   \n",
       "11  -11.056822   11.016819   14.420398          -16.829655   \n",
       "12  441.559954 -823.507585  476.351812              -200.0   \n",
       "13    0.113419   -6.235829    0.207506           -3.633004   \n",
       "14    29.00908  -28.197252   12.281439          -11.074747   \n",
       "15    15.51017  -17.725066   48.242652          -17.370662   \n",
       "\n",
       "   UCL[nonparam_boots] LCL[nonparam_stratified_boots]  \\\n",
       "0            -6.073176                         -200.0   \n",
       "1             0.000008                       -5.58638   \n",
       "2            29.050464                     -55.345204   \n",
       "3            32.228856                     -22.535263   \n",
       "4           -11.377163                         -200.0   \n",
       "5             0.000023                      -8.100437   \n",
       "6            53.358982                     -56.944076   \n",
       "7            32.542637                     -31.979314   \n",
       "8            -11.52958                         -200.0   \n",
       "9             0.000006                      -3.252504   \n",
       "10            9.058734                     -10.917826   \n",
       "11           17.473025                     -17.148796   \n",
       "12          -20.578777                         -200.0   \n",
       "13            0.000017                      -3.647139   \n",
       "14           10.755288                      -11.07049   \n",
       "15           17.734121                     -17.426096   \n",
       "\n",
       "   UCL[nonparam_stratified_boots] LCL[parametric_boots] UCL[parametric_boots]  \n",
       "0                       -6.054671                -200.0             -6.469164  \n",
       "1                        0.000011             -6.842527              0.000005  \n",
       "2                       28.169346            -65.983013             36.338077  \n",
       "3                       32.150612            -25.709363             32.237553  \n",
       "4                      -11.291955           -171.351895            -10.345907  \n",
       "5                        0.000039             -5.479493              0.000006  \n",
       "6                       54.305137            -59.586185              0.000111  \n",
       "7                       32.429343              -0.00053             28.444294  \n",
       "8                      -11.662121                -200.0            -13.819941  \n",
       "9                        0.000004             -9.281802              0.000033  \n",
       "10                       9.733296            -11.884471              8.003786  \n",
       "11                      17.244573            -16.781905             19.047468  \n",
       "12                     -21.084449                -200.0             -30.30607  \n",
       "13                       0.000013              -3.93403              0.000002  \n",
       "14                      10.754765            -12.071208             12.905614  \n",
       "15                      17.759353            -19.235026             18.703335  "
      ]
     },
     "execution_count": 86,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# convert to long format by exploding the pars, SE, LCL, UCL, beta columns\n",
    "cols_CI = df_pars_CI.columns[df_pars_CI.columns.str.startswith((\"LCL\", \"UCL\"))].tolist()\n",
    "cols = [\"idx\", \"beta\", \"pars\", \"SE\"] + cols_CI\n",
    "\n",
    "df_pars_CI_long = df_pars_CI.explode(cols, ignore_index=True)\n",
    "df_pars_CI_long"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "base (3.12.3)",
   "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": "3.12.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}