{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "5d0da7f5",
   "metadata": {},
   "source": [
    "# Multiverse integration\n",
    "\n",
    "Comet includes multiverse integration methods based on the weighting schemes introduced in:\n",
    "\n",
    "> Cantone, G. G., & Tomaselli, V. (2025). Characterisation and calibration of multiversal methods.  \n",
    "> Advances in Data Analysis and Classification, 19(4), 989-1021. https://doi.org/10.1007/s11634-024-00610-9\n",
    "\n",
    "For illustration and validation purposes, we use the \"unidentified multiverse\" example from the paper, which was introduced in a previous paper by Del Giudice and Gangestad (2021). https://doi.org/10.6084/m9.figshare.12089736"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "9999bfd8",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Removed 16 out of 256 universes:\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Universe</th>\n",
       "      <th>Decision 1</th>\n",
       "      <th>Value 1</th>\n",
       "      <th>Decision 2</th>\n",
       "      <th>Value 2</th>\n",
       "      <th>Decision 3</th>\n",
       "      <th>Value 3</th>\n",
       "      <th>Decision 4</th>\n",
       "      <th>Value 4</th>\n",
       "      <th>Decision 5</th>\n",
       "      <th>Value 5</th>\n",
       "      <th>Decision 6</th>\n",
       "      <th>Value 6</th>\n",
       "      <th>Decision 7</th>\n",
       "      <th>Value 7</th>\n",
       "      <th>Decision 8</th>\n",
       "      <th>Value 8</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Universe_1</td>\n",
       "      <td>X1</td>\n",
       "      <td>0</td>\n",
       "      <td>X2</td>\n",
       "      <td>0</td>\n",
       "      <td>X3</td>\n",
       "      <td>0</td>\n",
       "      <td>X4</td>\n",
       "      <td>1</td>\n",
       "      <td>ANTE</td>\n",
       "      <td>0</td>\n",
       "      <td>MEDI</td>\n",
       "      <td>0</td>\n",
       "      <td>COLL</td>\n",
       "      <td>0</td>\n",
       "      <td>CONF</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Universe_2</td>\n",
       "      <td>X1</td>\n",
       "      <td>0</td>\n",
       "      <td>X2</td>\n",
       "      <td>0</td>\n",
       "      <td>X3</td>\n",
       "      <td>0</td>\n",
       "      <td>X4</td>\n",
       "      <td>1</td>\n",
       "      <td>ANTE</td>\n",
       "      <td>0</td>\n",
       "      <td>MEDI</td>\n",
       "      <td>0</td>\n",
       "      <td>COLL</td>\n",
       "      <td>0</td>\n",
       "      <td>CONF</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Universe_3</td>\n",
       "      <td>X1</td>\n",
       "      <td>0</td>\n",
       "      <td>X2</td>\n",
       "      <td>0</td>\n",
       "      <td>X3</td>\n",
       "      <td>0</td>\n",
       "      <td>X4</td>\n",
       "      <td>1</td>\n",
       "      <td>ANTE</td>\n",
       "      <td>0</td>\n",
       "      <td>MEDI</td>\n",
       "      <td>0</td>\n",
       "      <td>COLL</td>\n",
       "      <td>1</td>\n",
       "      <td>CONF</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Universe_4</td>\n",
       "      <td>X1</td>\n",
       "      <td>0</td>\n",
       "      <td>X2</td>\n",
       "      <td>0</td>\n",
       "      <td>X3</td>\n",
       "      <td>0</td>\n",
       "      <td>X4</td>\n",
       "      <td>1</td>\n",
       "      <td>ANTE</td>\n",
       "      <td>0</td>\n",
       "      <td>MEDI</td>\n",
       "      <td>0</td>\n",
       "      <td>COLL</td>\n",
       "      <td>1</td>\n",
       "      <td>CONF</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Universe_5</td>\n",
       "      <td>X1</td>\n",
       "      <td>0</td>\n",
       "      <td>X2</td>\n",
       "      <td>0</td>\n",
       "      <td>X3</td>\n",
       "      <td>0</td>\n",
       "      <td>X4</td>\n",
       "      <td>1</td>\n",
       "      <td>ANTE</td>\n",
       "      <td>0</td>\n",
       "      <td>MEDI</td>\n",
       "      <td>1</td>\n",
       "      <td>COLL</td>\n",
       "      <td>0</td>\n",
       "      <td>CONF</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>235</th>\n",
       "      <td>Universe_236</td>\n",
       "      <td>X1</td>\n",
       "      <td>1</td>\n",
       "      <td>X2</td>\n",
       "      <td>1</td>\n",
       "      <td>X3</td>\n",
       "      <td>1</td>\n",
       "      <td>X4</td>\n",
       "      <td>1</td>\n",
       "      <td>ANTE</td>\n",
       "      <td>1</td>\n",
       "      <td>MEDI</td>\n",
       "      <td>0</td>\n",
       "      <td>COLL</td>\n",
       "      <td>1</td>\n",
       "      <td>CONF</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>236</th>\n",
       "      <td>Universe_237</td>\n",
       "      <td>X1</td>\n",
       "      <td>1</td>\n",
       "      <td>X2</td>\n",
       "      <td>1</td>\n",
       "      <td>X3</td>\n",
       "      <td>1</td>\n",
       "      <td>X4</td>\n",
       "      <td>1</td>\n",
       "      <td>ANTE</td>\n",
       "      <td>1</td>\n",
       "      <td>MEDI</td>\n",
       "      <td>1</td>\n",
       "      <td>COLL</td>\n",
       "      <td>0</td>\n",
       "      <td>CONF</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>237</th>\n",
       "      <td>Universe_238</td>\n",
       "      <td>X1</td>\n",
       "      <td>1</td>\n",
       "      <td>X2</td>\n",
       "      <td>1</td>\n",
       "      <td>X3</td>\n",
       "      <td>1</td>\n",
       "      <td>X4</td>\n",
       "      <td>1</td>\n",
       "      <td>ANTE</td>\n",
       "      <td>1</td>\n",
       "      <td>MEDI</td>\n",
       "      <td>1</td>\n",
       "      <td>COLL</td>\n",
       "      <td>0</td>\n",
       "      <td>CONF</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>238</th>\n",
       "      <td>Universe_239</td>\n",
       "      <td>X1</td>\n",
       "      <td>1</td>\n",
       "      <td>X2</td>\n",
       "      <td>1</td>\n",
       "      <td>X3</td>\n",
       "      <td>1</td>\n",
       "      <td>X4</td>\n",
       "      <td>1</td>\n",
       "      <td>ANTE</td>\n",
       "      <td>1</td>\n",
       "      <td>MEDI</td>\n",
       "      <td>1</td>\n",
       "      <td>COLL</td>\n",
       "      <td>1</td>\n",
       "      <td>CONF</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>239</th>\n",
       "      <td>Universe_240</td>\n",
       "      <td>X1</td>\n",
       "      <td>1</td>\n",
       "      <td>X2</td>\n",
       "      <td>1</td>\n",
       "      <td>X3</td>\n",
       "      <td>1</td>\n",
       "      <td>X4</td>\n",
       "      <td>1</td>\n",
       "      <td>ANTE</td>\n",
       "      <td>1</td>\n",
       "      <td>MEDI</td>\n",
       "      <td>1</td>\n",
       "      <td>COLL</td>\n",
       "      <td>1</td>\n",
       "      <td>CONF</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>240 rows × 17 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "         Universe Decision 1  Value 1 Decision 2  Value 2 Decision 3  Value 3  \\\n",
       "0      Universe_1         X1        0         X2        0         X3        0   \n",
       "1      Universe_2         X1        0         X2        0         X3        0   \n",
       "2      Universe_3         X1        0         X2        0         X3        0   \n",
       "3      Universe_4         X1        0         X2        0         X3        0   \n",
       "4      Universe_5         X1        0         X2        0         X3        0   \n",
       "..            ...        ...      ...        ...      ...        ...      ...   \n",
       "235  Universe_236         X1        1         X2        1         X3        1   \n",
       "236  Universe_237         X1        1         X2        1         X3        1   \n",
       "237  Universe_238         X1        1         X2        1         X3        1   \n",
       "238  Universe_239         X1        1         X2        1         X3        1   \n",
       "239  Universe_240         X1        1         X2        1         X3        1   \n",
       "\n",
       "    Decision 4  Value 4 Decision 5  Value 5 Decision 6  Value 6 Decision 7  \\\n",
       "0           X4        1       ANTE        0       MEDI        0       COLL   \n",
       "1           X4        1       ANTE        0       MEDI        0       COLL   \n",
       "2           X4        1       ANTE        0       MEDI        0       COLL   \n",
       "3           X4        1       ANTE        0       MEDI        0       COLL   \n",
       "4           X4        1       ANTE        0       MEDI        1       COLL   \n",
       "..         ...      ...        ...      ...        ...      ...        ...   \n",
       "235         X4        1       ANTE        1       MEDI        0       COLL   \n",
       "236         X4        1       ANTE        1       MEDI        1       COLL   \n",
       "237         X4        1       ANTE        1       MEDI        1       COLL   \n",
       "238         X4        1       ANTE        1       MEDI        1       COLL   \n",
       "239         X4        1       ANTE        1       MEDI        1       COLL   \n",
       "\n",
       "     Value 7 Decision 8  Value 8  \n",
       "0          0       CONF        0  \n",
       "1          0       CONF        1  \n",
       "2          1       CONF        0  \n",
       "3          1       CONF        1  \n",
       "4          0       CONF        0  \n",
       "..       ...        ...      ...  \n",
       "235        1       CONF        1  \n",
       "236        0       CONF        0  \n",
       "237        0       CONF        1  \n",
       "238        1       CONF        0  \n",
       "239        1       CONF        1  \n",
       "\n",
       "[240 rows x 17 columns]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Starting multiverse analysis for all universes...\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "5f0c6f7531ef48968761e0b826a304cb",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Performing multiverse analysis::   0%|          | 0/240 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The multiverse analysis completed without any errors.\n"
     ]
    }
   ],
   "source": [
    "from comet.multiverse import Multiverse\n",
    "\n",
    "forking_paths = {\n",
    "    \"X1\": [0, 1],\n",
    "    \"X2\": [0, 1],\n",
    "    \"X3\": [0, 1],\n",
    "    \"X4\": [0, 1],\n",
    "    \"ANTE\": [0, 1],\n",
    "    \"MEDI\": [0, 1],\n",
    "    \"COLL\": [0, 1],\n",
    "    \"CONF\": [0, 1],\n",
    "}\n",
    "\n",
    "def analysis_template():\n",
    "    import comet\n",
    "    import statsmodels.formula.api as smf\n",
    "\n",
    "    data = comet.utils.load_example(\"cantone\")\n",
    "    \n",
    "    # Build composite predictor\n",
    "    data[\"pred\"] = 0.0\n",
    "    if {{X1}}:\n",
    "        data[\"pred\"] += data[\"BM1\"]\n",
    "    if {{X2}}:\n",
    "        data[\"pred\"] += data[\"BM2\"]\n",
    "    if {{X3}}:\n",
    "        data[\"pred\"] += data[\"BM3\"]\n",
    "    if {{X4}}:\n",
    "        data[\"pred\"] += data[\"BM4\"]\n",
    "\n",
    "    # Build covariates\n",
    "    cov = \"\"\n",
    "    if {{ANTE}}:\n",
    "        cov += \" + genotype\"\n",
    "    if {{MEDI}}:\n",
    "        cov += \" + pain\"\n",
    "    if {{COLL}}:\n",
    "        cov += \" + fatigue\"\n",
    "    if {{CONF}}:\n",
    "        cov += \" + age\"\n",
    "\n",
    "    # Regression model\n",
    "    formula = \"depression ~ pred\" + cov\n",
    "    fit_res = smf.ols(formula=formula, data=data).fit()\n",
    "\n",
    "    beta_raw = float(fit_res.params[\"pred\"])\n",
    "    sd_y = float(data[\"depression\"].std())\n",
    "    sd_x = float(data[\"pred\"].std())\n",
    "    beta_std = beta_raw * (sd_x / sd_y)\n",
    "\n",
    "    out = {\n",
    "        \"formula\": formula,\n",
    "        \"beta\": float(beta_std),\n",
    "        \"beta_raw\": float(beta_raw),\n",
    "        \"p_value\": float(fit_res.pvalues[\"pred\"]),\n",
    "    }\n",
    "\n",
    "    comet.utils.save_universe_results(out)\n",
    "\n",
    "config = {\n",
    "    \"remove\": [\n",
    "        [{\"X1\": 0}, {\"X2\": 0}, {\"X3\": 0}, {\"X4\": 0}],\n",
    "    ]\n",
    "}\n",
    "\n",
    "mverse = Multiverse(name=\"example_mv_integration\")\n",
    "mverse.create(analysis_template, forking_paths, config)\n",
    "mverse.summary()\n",
    "mverse.run(parallel=16)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "951fb475",
   "metadata": {},
   "source": [
    "We visualise the multiverse results:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "ba791560",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 700x900 with 10 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "mverse.multiverse_plot(measure=\"beta\", n_bins=20, sig_col=\"p_value\", baseline=0, figsize=(7,9))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7b238452",
   "metadata": {},
   "source": [
    "Comet currently contains two methods for integrating over the multiverse:\n",
    "\n",
    "`uniform`\n",
    "- All universes are weighted equally (the estimate is just the plain mean/median across universes). This corresponds to the density function as plotted in the multiverse plot above.\n",
    "\n",
    "`mli` (Maximum Local Influence)\n",
    "- Each universe is compared with its nearest neighbours (specifications differing by a single decision). Universes whose estimate barely changes when one decision is flipped are weighted more, down-weighting results that hinge on a single arbitrary choice.\n",
    "\n",
    "As an example, we can use the `.integrate()` method to simply get an integrated measure over all universes. For example, using uniform weighting and mean aggregation:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "177e98c0",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Mean beta: -0.0066\n"
     ]
    }
   ],
   "source": [
    "estimate, weights = mverse.integrate(measure=\"beta\", method=\"uniform\", type=\"mean\")\n",
    "print(f\"Mean beta: {estimate:.4f}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "33dbe1c3",
   "metadata": {},
   "source": [
    "We can also use the `compare_integration` and `plot_integration` convenience functions to get a summary for both integration methods."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "d3c42976",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Scheme</th>\n",
       "      <th>median</th>\n",
       "      <th>mean</th>\n",
       "      <th>gini_w</th>\n",
       "      <th>err_median</th>\n",
       "      <th>err_mean</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Uniform</td>\n",
       "      <td>-0.0210</td>\n",
       "      <td>-0.0066</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.221</td>\n",
       "      <td>0.207</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>MLI</td>\n",
       "      <td>-0.0328</td>\n",
       "      <td>-0.0144</td>\n",
       "      <td>0.243</td>\n",
       "      <td>0.233</td>\n",
       "      <td>0.214</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    Scheme  median    mean  gini_w  err_median  err_mean\n",
       "0  Uniform -0.0210 -0.0066   0.000       0.221     0.207\n",
       "1      MLI -0.0328 -0.0144   0.243       0.233     0.214"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "table, weights = mverse.compare_integration(measure=\"beta\", true_value=0.2)\n",
    "table"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "91f5e652",
   "metadata": {},
   "source": [
    "The columns are, for each weighting scheme:\n",
    "\n",
    "- `median` / `mean`: the integrated estimate (weighted median/mean of `beta`)\n",
    "- `gini_w`: how concentrated the weights are (0 = all equal, → 1 = dominated by a few universes).\n",
    "\n",
    "If `true_value` is passed, additional `err_*` columns give the absolute error of each estimate against it. Usually, this is not available in real multiverses, but we can include it here as we have the ground truth from the simulation."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "daad2d7a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 700x300 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "mverse.plot_integration(measure=\"beta\", weights=weights, true_value=0.2)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "comet",
   "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": "3.13.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}