{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "78bce0e3",
   "metadata": {},
   "source": [
    "# Lesson 5: Why Hierarchical Modelling Beats Maximum Likelihood\n",
    "\n",
    "## The problem with individual fitting\n",
    "\n",
    "Maximum-likelihood estimation (MLE) and MAP estimation fit each participant independently.\n",
    "With hundreds of trials per subject this works fine, but in practice most psychophysics\n",
    "experiments give you **100--250 trials per condition**.  At those trial counts individual\n",
    "MLE/MAP estimates are:\n",
    "\n",
    "- **Noisy** — refit the same subject on a different random half of their data and you get\n",
    "  a substantially different answer.\n",
    "- **Biased at the boundaries** — with few trials, the likelihood surface is broad and\n",
    "  the optimiser can land at extreme values.  Noise parameters ($\\nu$) may converge to\n",
    "  near-zero (the model \"explains\" every trial perfectly by overfitting) or explode to\n",
    "  very large values (flat psychometric curve, no discrimination at all).  Lapse-rate\n",
    "  parameters can rail at 0 or 1.  These boundary estimates are not meaningful — they\n",
    "  reflect the instability of the optimisation, not the participant's true noise level.\n",
    "- **Worse for complex models** — every additional parameter increases the volume of\n",
    "  the parameter space that the optimiser must search.  The most theoretically interesting\n",
    "  models (KLW with prior $\\sigma_0$, FlexibleNoise with spline coefficients) are exactly\n",
    "  the ones that suffer most, because they have more parameters per subject and more\n",
    "  opportunities for the likelihood to be flat.\n",
    "\n",
    "**Hierarchical Bayesian estimation** avoids all three problems.  By sharing statistical\n",
    "strength across participants, the group prior acts as an *adaptive regulariser*: subjects\n",
    "with extreme or unreliable data are pulled toward the population, while well-identified\n",
    "subjects are left alone.\n",
    "\n",
    "## Empirical demonstration: split-half reliability\n",
    "\n",
    "The gold standard for measurement quality is **split-half reliability**: split each\n",
    "participant's trials into two random halves, fit each half separately, and correlate the\n",
    "parameter estimates across halves.  A reliable method produces the same answer from both\n",
    "halves; an unreliable one does not.\n",
    "\n",
    "We compare:\n",
    "\n",
    "| Method | What it does | Regularisation |\n",
    "|--------|-------------|----------------|\n",
    "| **MLE** | `model.fit_map_individual(flat_prior=True)` — each subject fitted alone with flat priors | None — pure maximum likelihood |\n",
    "| **Hierarchical Bayes** | `model.sample()` with `hierarchical=True` — full MCMC posterior | Adaptive group prior, posterior averaging |\n",
    "\n",
    "at increasing trial counts (25, 50, 108 per half).  The Garcia et al. magnitude\n",
    "data have 216 trials per subject, so 108 per half is the natural split."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "0cc30d54",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-04-10T06:20:46.193800Z",
     "iopub.status.busy": "2026-04-10T06:20:46.193612Z",
     "iopub.status.idle": "2026-04-10T06:20:50.770550Z",
     "shell.execute_reply": "2026-04-10T06:20:50.770139Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Subjects: 64,  trials per subject: 216\n",
      "Half A: 6912 trials,  Half B: 6912 trials\n"
     ]
    }
   ],
   "source": [
    "import warnings; warnings.filterwarnings('ignore')\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "from bauer.utils.data import load_garcia2022\n",
    "from bauer.models import MagnitudeComparisonModel\n",
    "\n",
    "df_mag = load_garcia2022(task='magnitude')\n",
    "subjects = df_mag.index.get_level_values('subject').unique()\n",
    "print(f\"Subjects: {len(subjects)},  trials per subject: \"\n",
    "      f\"{df_mag.groupby(level='subject').size().iloc[0]}\")\n",
    "\n",
    "# ── Split each subject's trials into two random halves ───────────────────────\n",
    "rng = np.random.default_rng(42)\n",
    "half_a_rows, half_b_rows = [], []\n",
    "for subj in subjects:\n",
    "    mask = df_mag.index.get_level_values('subject') == subj\n",
    "    subj_iloc = np.where(mask)[0]\n",
    "    perm = rng.permutation(len(subj_iloc))\n",
    "    mid = len(subj_iloc) // 2\n",
    "    half_a_rows.extend(subj_iloc[perm[:mid]])\n",
    "    half_b_rows.extend(subj_iloc[perm[mid:]])\n",
    "\n",
    "half_a_full = df_mag.iloc[half_a_rows]\n",
    "half_b_full = df_mag.iloc[half_b_rows]\n",
    "print(f\"Half A: {len(half_a_full)} trials,  Half B: {len(half_b_full)} trials\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "86bf24ab",
   "metadata": {},
   "source": [
    "## Split-half analysis: 10 random splits $\\times$ 5 trial counts\n",
    "\n",
    "For each trial count (25, 50, 108) we:\n",
    "1. Randomly split each subject's data into two halves\n",
    "2. Subsample *k* trials per half\n",
    "3. Fit MLE and hierarchical Bayes on each half\n",
    "4. Correlate the parameter estimates across halves\n",
    "\n",
    "We repeat this 10 times with different random splits to get a stable estimate of\n",
    "reliability.  Three correlation metrics are shown: Spearman $\\rho$, Pearson $r$,\n",
    "and $R^2$."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "9f59d84f",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-04-10T06:20:50.772792Z",
     "iopub.status.busy": "2026-04-10T06:20:50.772571Z",
     "iopub.status.idle": "2026-04-10T07:52:03.126311Z",
     "shell.execute_reply": "2026-04-10T07:52:03.125979Z"
    }
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Initializing NUTS using jitter+adapt_diag...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Multiprocess sampling (2 chains in 2 jobs)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "NUTS: [n1_evidence_sd_mu_untransformed, n1_evidence_sd_sd, n1_evidence_sd_offset, n2_evidence_sd_mu_untransformed, n2_evidence_sd_sd, n2_evidence_sd_offset]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Sampling 2 chains for 500 tune and 500 draw iterations (1_000 + 1_000 draws total) took 5 seconds.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "There was 1 divergence after tuning. Increase `target_accept` or reparameterize.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "We recommend running at least 4 chains for robust computation of convergence diagnostics\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "The rhat statistic is larger than 1.01 for some parameters. This indicates problems during sampling. See https://arxiv.org/abs/1903.08008 for details\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "The effective sample size per chain is smaller than 100 for some parameters.  A higher number is needed for reliable rhat and ess computation. See https://arxiv.org/abs/1903.08008 for details\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Initializing NUTS using jitter+adapt_diag...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Multiprocess sampling (2 chains in 2 jobs)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "NUTS: [n1_evidence_sd_mu_untransformed, n1_evidence_sd_sd, n1_evidence_sd_offset, n2_evidence_sd_mu_untransformed, n2_evidence_sd_sd, n2_evidence_sd_offset]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Sampling 2 chains for 500 tune and 500 draw iterations (1_000 + 1_000 draws total) took 5 seconds.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "There were 6 divergences after tuning. Increase `target_accept` or reparameterize.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "We recommend running at least 4 chains for robust computation of convergence diagnostics\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "The rhat statistic is larger than 1.01 for some parameters. This indicates problems during sampling. See https://arxiv.org/abs/1903.08008 for details\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "The effective sample size per chain is smaller than 100 for some parameters.  A higher number is needed for reliable rhat and ess computation. See https://arxiv.org/abs/1903.08008 for details\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Initializing NUTS using jitter+adapt_diag...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Multiprocess sampling (2 chains in 2 jobs)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "NUTS: [n1_evidence_sd_mu_untransformed, n1_evidence_sd_sd, n1_evidence_sd_offset, n2_evidence_sd_mu_untransformed, n2_evidence_sd_sd, n2_evidence_sd_offset]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Sampling 2 chains for 500 tune and 500 draw iterations (1_000 + 1_000 draws total) took 9 seconds.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "There were 16 divergences after tuning. Increase `target_accept` or reparameterize.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "We recommend running at least 4 chains for robust computation of convergence diagnostics\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "The rhat statistic is larger than 1.01 for some parameters. This indicates problems during sampling. See https://arxiv.org/abs/1903.08008 for details\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Initializing NUTS using jitter+adapt_diag...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Multiprocess sampling (2 chains in 2 jobs)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "NUTS: [n1_evidence_sd_mu_untransformed, n1_evidence_sd_sd, n1_evidence_sd_offset, n2_evidence_sd_mu_untransformed, n2_evidence_sd_sd, n2_evidence_sd_offset]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Sampling 2 chains for 500 tune and 500 draw iterations (1_000 + 1_000 draws total) took 13 seconds.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "There were 9 divergences after tuning. Increase `target_accept` or reparameterize.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "We recommend running at least 4 chains for robust computation of convergence diagnostics\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "The rhat statistic is larger than 1.01 for some parameters. This indicates problems during sampling. See https://arxiv.org/abs/1903.08008 for details\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Initializing NUTS using jitter+adapt_diag...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Multiprocess sampling (2 chains in 2 jobs)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "NUTS: [n1_evidence_sd_mu_untransformed, n1_evidence_sd_sd, n1_evidence_sd_offset, n2_evidence_sd_mu_untransformed, n2_evidence_sd_sd, n2_evidence_sd_offset]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Sampling 2 chains for 500 tune and 500 draw iterations (1_000 + 1_000 draws total) took 28 seconds.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "There were 5 divergences after tuning. Increase `target_accept` or reparameterize.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "We recommend running at least 4 chains for robust computation of convergence diagnostics\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "The rhat statistic is larger than 1.01 for some parameters. This indicates problems during sampling. See https://arxiv.org/abs/1903.08008 for details\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "The effective sample size per chain is smaller than 100 for some parameters.  A higher number is needed for reliable rhat and ess computation. See https://arxiv.org/abs/1903.08008 for details\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Initializing NUTS using jitter+adapt_diag...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Multiprocess sampling (2 chains in 2 jobs)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "NUTS: [n1_evidence_sd_mu_untransformed, n1_evidence_sd_sd, n1_evidence_sd_offset, n2_evidence_sd_mu_untransformed, n2_evidence_sd_sd, n2_evidence_sd_offset]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Sampling 2 chains for 500 tune and 500 draw iterations (1_000 + 1_000 draws total) took 22 seconds.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "There were 2 divergences after tuning. Increase `target_accept` or reparameterize.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "We recommend running at least 4 chains for robust computation of convergence diagnostics\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "The rhat statistic is larger than 1.01 for some parameters. This indicates problems during sampling. See https://arxiv.org/abs/1903.08008 for details\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Split 1/3 done\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Initializing NUTS using jitter+adapt_diag...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Multiprocess sampling (2 chains in 2 jobs)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "NUTS: [n1_evidence_sd_mu_untransformed, n1_evidence_sd_sd, n1_evidence_sd_offset, n2_evidence_sd_mu_untransformed, n2_evidence_sd_sd, n2_evidence_sd_offset]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Sampling 2 chains for 500 tune and 500 draw iterations (1_000 + 1_000 draws total) took 4 seconds.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "There was 1 divergence after tuning. Increase `target_accept` or reparameterize.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "We recommend running at least 4 chains for robust computation of convergence diagnostics\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "The rhat statistic is larger than 1.01 for some parameters. This indicates problems during sampling. See https://arxiv.org/abs/1903.08008 for details\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Initializing NUTS using jitter+adapt_diag...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Multiprocess sampling (2 chains in 2 jobs)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "NUTS: [n1_evidence_sd_mu_untransformed, n1_evidence_sd_sd, n1_evidence_sd_offset, n2_evidence_sd_mu_untransformed, n2_evidence_sd_sd, n2_evidence_sd_offset]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Sampling 2 chains for 500 tune and 500 draw iterations (1_000 + 1_000 draws total) took 5 seconds.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "There were 4 divergences after tuning. Increase `target_accept` or reparameterize.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "We recommend running at least 4 chains for robust computation of convergence diagnostics\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "The rhat statistic is larger than 1.01 for some parameters. This indicates problems during sampling. See https://arxiv.org/abs/1903.08008 for details\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "The effective sample size per chain is smaller than 100 for some parameters.  A higher number is needed for reliable rhat and ess computation. See https://arxiv.org/abs/1903.08008 for details\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Initializing NUTS using jitter+adapt_diag...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Multiprocess sampling (2 chains in 2 jobs)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "NUTS: [n1_evidence_sd_mu_untransformed, n1_evidence_sd_sd, n1_evidence_sd_offset, n2_evidence_sd_mu_untransformed, n2_evidence_sd_sd, n2_evidence_sd_offset]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Sampling 2 chains for 500 tune and 500 draw iterations (1_000 + 1_000 draws total) took 28 seconds.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "There was 1 divergence after tuning. Increase `target_accept` or reparameterize.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "We recommend running at least 4 chains for robust computation of convergence diagnostics\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "The rhat statistic is larger than 1.01 for some parameters. This indicates problems during sampling. See https://arxiv.org/abs/1903.08008 for details\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Initializing NUTS using jitter+adapt_diag...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Multiprocess sampling (2 chains in 2 jobs)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "NUTS: [n1_evidence_sd_mu_untransformed, n1_evidence_sd_sd, n1_evidence_sd_offset, n2_evidence_sd_mu_untransformed, n2_evidence_sd_sd, n2_evidence_sd_offset]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Sampling 2 chains for 500 tune and 500 draw iterations (1_000 + 1_000 draws total) took 10 seconds.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "There were 34 divergences after tuning. Increase `target_accept` or reparameterize.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "We recommend running at least 4 chains for robust computation of convergence diagnostics\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "The rhat statistic is larger than 1.01 for some parameters. This indicates problems during sampling. See https://arxiv.org/abs/1903.08008 for details\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "The effective sample size per chain is smaller than 100 for some parameters.  A higher number is needed for reliable rhat and ess computation. See https://arxiv.org/abs/1903.08008 for details\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Initializing NUTS using jitter+adapt_diag...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Multiprocess sampling (2 chains in 2 jobs)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "NUTS: [n1_evidence_sd_mu_untransformed, n1_evidence_sd_sd, n1_evidence_sd_offset, n2_evidence_sd_mu_untransformed, n2_evidence_sd_sd, n2_evidence_sd_offset]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Sampling 2 chains for 500 tune and 500 draw iterations (1_000 + 1_000 draws total) took 30 seconds.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "There were 4 divergences after tuning. Increase `target_accept` or reparameterize.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "We recommend running at least 4 chains for robust computation of convergence diagnostics\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "The rhat statistic is larger than 1.01 for some parameters. This indicates problems during sampling. See https://arxiv.org/abs/1903.08008 for details\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "The effective sample size per chain is smaller than 100 for some parameters.  A higher number is needed for reliable rhat and ess computation. See https://arxiv.org/abs/1903.08008 for details\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Initializing NUTS using jitter+adapt_diag...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Multiprocess sampling (2 chains in 2 jobs)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "NUTS: [n1_evidence_sd_mu_untransformed, n1_evidence_sd_sd, n1_evidence_sd_offset, n2_evidence_sd_mu_untransformed, n2_evidence_sd_sd, n2_evidence_sd_offset]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Sampling 2 chains for 500 tune and 500 draw iterations (1_000 + 1_000 draws total) took 22 seconds.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "There were 5 divergences after tuning. Increase `target_accept` or reparameterize.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "We recommend running at least 4 chains for robust computation of convergence diagnostics\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "The rhat statistic is larger than 1.01 for some parameters. This indicates problems during sampling. See https://arxiv.org/abs/1903.08008 for details\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "The effective sample size per chain is smaller than 100 for some parameters.  A higher number is needed for reliable rhat and ess computation. See https://arxiv.org/abs/1903.08008 for details\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Split 2/3 done\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Initializing NUTS using jitter+adapt_diag...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Multiprocess sampling (2 chains in 2 jobs)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "NUTS: [n1_evidence_sd_mu_untransformed, n1_evidence_sd_sd, n1_evidence_sd_offset, n2_evidence_sd_mu_untransformed, n2_evidence_sd_sd, n2_evidence_sd_offset]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Sampling 2 chains for 500 tune and 500 draw iterations (1_000 + 1_000 draws total) took 5 seconds.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "There were 4 divergences after tuning. Increase `target_accept` or reparameterize.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "We recommend running at least 4 chains for robust computation of convergence diagnostics\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "The rhat statistic is larger than 1.01 for some parameters. This indicates problems during sampling. See https://arxiv.org/abs/1903.08008 for details\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "The effective sample size per chain is smaller than 100 for some parameters.  A higher number is needed for reliable rhat and ess computation. See https://arxiv.org/abs/1903.08008 for details\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Initializing NUTS using jitter+adapt_diag...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Multiprocess sampling (2 chains in 2 jobs)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "NUTS: [n1_evidence_sd_mu_untransformed, n1_evidence_sd_sd, n1_evidence_sd_offset, n2_evidence_sd_mu_untransformed, n2_evidence_sd_sd, n2_evidence_sd_offset]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Sampling 2 chains for 500 tune and 500 draw iterations (1_000 + 1_000 draws total) took 5 seconds.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "There were 8 divergences after tuning. Increase `target_accept` or reparameterize.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "We recommend running at least 4 chains for robust computation of convergence diagnostics\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "The rhat statistic is larger than 1.01 for some parameters. This indicates problems during sampling. See https://arxiv.org/abs/1903.08008 for details\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Initializing NUTS using jitter+adapt_diag...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Multiprocess sampling (2 chains in 2 jobs)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "NUTS: [n1_evidence_sd_mu_untransformed, n1_evidence_sd_sd, n1_evidence_sd_offset, n2_evidence_sd_mu_untransformed, n2_evidence_sd_sd, n2_evidence_sd_offset]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Sampling 2 chains for 500 tune and 500 draw iterations (1_000 + 1_000 draws total) took 10 seconds.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "There were 9 divergences after tuning. Increase `target_accept` or reparameterize.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "We recommend running at least 4 chains for robust computation of convergence diagnostics\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "The rhat statistic is larger than 1.01 for some parameters. This indicates problems during sampling. See https://arxiv.org/abs/1903.08008 for details\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "The effective sample size per chain is smaller than 100 for some parameters.  A higher number is needed for reliable rhat and ess computation. See https://arxiv.org/abs/1903.08008 for details\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Initializing NUTS using jitter+adapt_diag...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Multiprocess sampling (2 chains in 2 jobs)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "NUTS: [n1_evidence_sd_mu_untransformed, n1_evidence_sd_sd, n1_evidence_sd_offset, n2_evidence_sd_mu_untransformed, n2_evidence_sd_sd, n2_evidence_sd_offset]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Sampling 2 chains for 500 tune and 500 draw iterations (1_000 + 1_000 draws total) took 10 seconds.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "There were 41 divergences after tuning. Increase `target_accept` or reparameterize.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "We recommend running at least 4 chains for robust computation of convergence diagnostics\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "The rhat statistic is larger than 1.01 for some parameters. This indicates problems during sampling. See https://arxiv.org/abs/1903.08008 for details\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "The effective sample size per chain is smaller than 100 for some parameters.  A higher number is needed for reliable rhat and ess computation. See https://arxiv.org/abs/1903.08008 for details\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Initializing NUTS using jitter+adapt_diag...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Multiprocess sampling (2 chains in 2 jobs)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "NUTS: [n1_evidence_sd_mu_untransformed, n1_evidence_sd_sd, n1_evidence_sd_offset, n2_evidence_sd_mu_untransformed, n2_evidence_sd_sd, n2_evidence_sd_offset]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Sampling 2 chains for 500 tune and 500 draw iterations (1_000 + 1_000 draws total) took 21 seconds.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "There were 87 divergences after tuning. Increase `target_accept` or reparameterize.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "We recommend running at least 4 chains for robust computation of convergence diagnostics\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "The rhat statistic is larger than 1.01 for some parameters. This indicates problems during sampling. See https://arxiv.org/abs/1903.08008 for details\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "The effective sample size per chain is smaller than 100 for some parameters.  A higher number is needed for reliable rhat and ess computation. See https://arxiv.org/abs/1903.08008 for details\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Initializing NUTS using jitter+adapt_diag...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Multiprocess sampling (2 chains in 2 jobs)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "NUTS: [n1_evidence_sd_mu_untransformed, n1_evidence_sd_sd, n1_evidence_sd_offset, n2_evidence_sd_mu_untransformed, n2_evidence_sd_sd, n2_evidence_sd_offset]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Sampling 2 chains for 500 tune and 500 draw iterations (1_000 + 1_000 draws total) took 21 seconds.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "There were 6 divergences after tuning. Increase `target_accept` or reparameterize.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "We recommend running at least 4 chains for robust computation of convergence diagnostics\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "The rhat statistic is larger than 1.01 for some parameters. This indicates problems during sampling. See https://arxiv.org/abs/1903.08008 for details\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "The effective sample size per chain is smaller than 100 for some parameters.  A higher number is needed for reliable rhat and ess computation. See https://arxiv.org/abs/1903.08008 for details\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Split 3/3 done\n",
      "\n",
      "Total fits: 54\n"
     ]
    }
   ],
   "source": [
    "from scipy.stats import pearsonr\n",
    "\n",
    "def subsample(df, k):\n",
    "    return df.groupby(level='subject').head(k)\n",
    "\n",
    "def split_data(df_mag, seed):\n",
    "    # Random split of each subject's trials into two halves\n",
    "    rng = np.random.default_rng(seed)\n",
    "    a_rows, b_rows = [], []\n",
    "    for subj in df_mag.index.get_level_values('subject').unique():\n",
    "        mask = df_mag.index.get_level_values('subject') == subj\n",
    "        iloc = np.where(mask)[0]\n",
    "        perm = rng.permutation(len(iloc))\n",
    "        mid = len(iloc) // 2\n",
    "        a_rows.extend(iloc[perm[:mid]])\n",
    "        b_rows.extend(iloc[perm[mid:]])\n",
    "    return df_mag.iloc[a_rows], df_mag.iloc[b_rows]\n",
    "\n",
    "def fit_mle(data):\n",
    "    model = MagnitudeComparisonModel(paradigm=data, fit_seperate_evidence_sd=True)\n",
    "    return model.fit_map_individual(data=data, flat_prior=True)\n",
    "\n",
    "def fit_hierarchical(data, draws=500, tune=500, chains=2):\n",
    "    model = MagnitudeComparisonModel(paradigm=data, fit_seperate_evidence_sd=True)\n",
    "    model.build_estimation_model(data=data, hierarchical=True)\n",
    "    idata = model.sample(draws=draws, tune=tune, chains=chains, progressbar=False)\n",
    "    n_subj = len(data.index.unique(level='subject'))\n",
    "    n1 = idata.posterior['n1_evidence_sd'].values.reshape(-1, n_subj).mean(0)\n",
    "    n2 = idata.posterior['n2_evidence_sd'].values.reshape(-1, n_subj).mean(0)\n",
    "    return pd.DataFrame({'n1_evidence_sd': n1, 'n2_evidence_sd': n2},\n",
    "                         index=pd.Index(data.index.unique(level='subject'), name='subject'))\n",
    "\n",
    "trial_counts = [25, 50, 108]\n",
    "n_splits = 3\n",
    "methods = {'MLE': fit_mle, 'Hierarchical Bayes': fit_hierarchical}\n",
    "results = []\n",
    "\n",
    "for split_i in range(n_splits):\n",
    "    half_a, half_b = split_data(df_mag, seed=split_i)\n",
    "    for k in trial_counts:\n",
    "        a_sub = subsample(half_a, k)\n",
    "        b_sub = subsample(half_b, k)\n",
    "        for method_name, fit_fn in methods.items():\n",
    "            est_a = fit_fn(a_sub)\n",
    "            est_b = fit_fn(b_sub)\n",
    "            # Also compute sum of both noise params\n",
    "            est_a['total_sd'] = est_a['n1_evidence_sd'] + est_a['n2_evidence_sd']\n",
    "            est_b['total_sd'] = est_b['n1_evidence_sd'] + est_b['n2_evidence_sd']\n",
    "            for param in ['n1_evidence_sd', 'n2_evidence_sd', 'total_sd']:\n",
    "                rho_p, _ = pearsonr(est_a[param], est_b[param])\n",
    "                # Count boundary-collapsed estimates (noise near zero)\n",
    "                n_zero_a = (est_a[param] < 0.01).sum()\n",
    "                n_zero_b = (est_b[param] < 0.01).sum()\n",
    "                results.append({\n",
    "                    'split': split_i, 'k': k, 'method': method_name,\n",
    "                    'parameter': param,\n",
    "                    'Pearson r': rho_p, 'R\\u00b2': rho_p**2,\n",
    "                    'n_collapsed': n_zero_a + n_zero_b,\n",
    "                })\n",
    "    print(f\"Split {split_i+1}/{n_splits} done\")\n",
    "\n",
    "results_df = pd.DataFrame(results)\n",
    "print(f\"\\nTotal fits: {len(results_df)}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "48c4ff21",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-04-10T07:52:03.131002Z",
     "iopub.status.busy": "2026-04-10T07:52:03.130864Z",
     "iopub.status.idle": "2026-04-10T07:52:03.738943Z",
     "shell.execute_reply": "2026-04-10T07:52:03.738671Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1500x700 with 6 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# ── Reliability vs trial count: Pearson r and R² x 3 parameters ──────────────\n",
    "metrics = ['Pearson r', 'R\\u00b2']\n",
    "params = ['n1_evidence_sd', 'n2_evidence_sd', 'total_sd']\n",
    "param_labels = {'n1_evidence_sd': '\\u03bd\\u2081 (first option)',\n",
    "                'n2_evidence_sd': '\\u03bd\\u2082 (second option)',\n",
    "                'total_sd': '\\u03bd\\u2081 + \\u03bd\\u2082 (total)'}\n",
    "pal = {'MLE': '#d73027', 'Hierarchical Bayes': '#4393c3'}\n",
    "\n",
    "fig, axes = plt.subplots(len(metrics), len(params),\n",
    "                          figsize=(5 * len(params), 3.5 * len(metrics)),\n",
    "                          sharex=True, sharey='row')\n",
    "\n",
    "for row, metric in enumerate(metrics):\n",
    "    for col, param in enumerate(params):\n",
    "        ax = axes[row, col]\n",
    "        sub = results_df[results_df['parameter'] == param]\n",
    "        for i_m, method in enumerate(['MLE', 'Hierarchical Bayes']):\n",
    "            d = sub[sub['method'] == method]\n",
    "            mean_by_k = d.groupby('k')[metric].mean()\n",
    "            se_by_k   = d.groupby('k')[metric].std() / np.sqrt(n_splits)\n",
    "            offset = (i_m - 0.5) * 1.5  # slight x-offset to avoid overlap\n",
    "            ax.errorbar(mean_by_k.index + offset, mean_by_k,\n",
    "                        yerr=1.96 * se_by_k,\n",
    "                        fmt='o-', lw=2, ms=6, capsize=4, capthick=1.5,\n",
    "                        color=pal[method], ecolor=pal[method], label=method)\n",
    "        ax.axhline(0, ls=':', c='gray', lw=1)\n",
    "        if row == 0:\n",
    "            ax.set_title(param_labels[param])\n",
    "        if row == len(metrics) - 1:\n",
    "            ax.set_xlabel('Trials per half')\n",
    "        if col == 0:\n",
    "            ax.set_ylabel(metric)\n",
    "        ax.legend(fontsize=8)\n",
    "        sns.despine(ax=ax)\n",
    "\n",
    "plt.suptitle(f'Split-half reliability (mean \\u00b1 95% CI over {n_splits} splits)',\n",
    "             fontsize=13, y=1.01)\n",
    "plt.tight_layout()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "60fcc7e4",
   "metadata": {},
   "source": [
    "### The boundary-collapse problem\n",
    "\n",
    "Why does MLE do so poorly on Pearson $r$ (which measures linear agreement) even when\n",
    "Spearman $\\rho$ (which only checks rank order) looks acceptable?  The answer is\n",
    "**boundary collapse**: with few trials, the MLE optimiser can push noise parameters\n",
    "$\\nu_1$ or $\\nu_2$ to near-zero, meaning the model \"explains\" every trial perfectly by\n",
    "overfitting the noise away.  These zero-estimates are meaningless — they do not reflect\n",
    "a participant with perfect perception, they reflect an optimiser that ran out of data.\n",
    "\n",
    "The plot below shows how many subjects (out of $N \\times 2$ halves) have $\\nu < 0.01$\n",
    "at each trial count:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "ef8434d0",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-04-10T07:52:03.740613Z",
     "iopub.status.busy": "2026-04-10T07:52:03.740495Z",
     "iopub.status.idle": "2026-04-10T07:52:03.922300Z",
     "shell.execute_reply": "2026-04-10T07:52:03.921956Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1400x400 with 3 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# ── Boundary collapse diagnostic ─────────────────────────────────────────────\n",
    "fig, axes = plt.subplots(1, 3, figsize=(14, 4), sharey=True)\n",
    "for col, param in enumerate(params):\n",
    "    ax = axes[col]\n",
    "    sub = results_df[results_df['parameter'] == param]\n",
    "    for i_m, method in enumerate(['MLE', 'Hierarchical Bayes']):\n",
    "        d = sub[sub['method'] == method]\n",
    "        mean_col = d.groupby('k')['n_collapsed'].mean()\n",
    "        se_col   = d.groupby('k')['n_collapsed'].std() / np.sqrt(n_splits)\n",
    "        offset = (i_m - 0.5) * 1.5\n",
    "        ax.errorbar(mean_col.index + offset, mean_col,\n",
    "                    yerr=1.96 * se_col,\n",
    "                    fmt='o-', lw=2, ms=6, capsize=4, capthick=1.5,\n",
    "                    color=pal[method], ecolor=pal[method], label=method)\n",
    "    ax.axhline(0, ls=':', c='gray', lw=1)\n",
    "    ax.set_title(param_labels[param])\n",
    "    ax.set_xlabel('Trials per half')\n",
    "    if col == 0:\n",
    "        ax.set_ylabel('Subjects with \\u03bd < 0.01\\n(out of 2 \\u00d7 N halves)')\n",
    "    ax.legend(fontsize=8)\n",
    "    sns.despine(ax=ax)\n",
    "\n",
    "plt.suptitle('Boundary collapse: MLE pushes noise to zero when data are scarce',\n",
    "             fontsize=12, y=1.02)\n",
    "plt.tight_layout()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a38a1cfb",
   "metadata": {},
   "source": [
    "## What this means in practice\n",
    "\n",
    "The plot shows what every psychophysics researcher should know but few textbooks\n",
    "make explicit:\n",
    "\n",
    "1. **Individual MAP/MLE is unreliable at typical trial counts.**  At 50--100 trials per\n",
    "   subject the split-half correlation can be near zero — the estimates are dominated by\n",
    "   sampling noise and tell you almost nothing about the participant.\n",
    "\n",
    "2. **Hierarchical Bayes is dramatically more reliable.**  The group prior acts as adaptive\n",
    "   regularisation: extreme/noisy estimates are pulled toward the group mean, which is\n",
    "   exactly what reduces split-half variability.  The posterior mean is a *shrinkage\n",
    "   estimator* — the optimal bias-variance trade-off.\n",
    "\n",
    "3. **The advantage is largest when you need it most** — at low trial counts and for\n",
    "   complex models with many parameters.  The KLW model (lesson 2) and the FlexibleNoise\n",
    "   model (lesson 4) have 4+ parameters per subject; individual MLE would be hopeless\n",
    "   without hundreds of trials, but hierarchical fitting works with the trial counts we\n",
    "   actually have.\n",
    "\n",
    "### Why MAP/MLE fails: the bias-variance trade-off\n",
    "\n",
    "MAP gives the single point with highest posterior density — for a flat prior this is\n",
    "just maximum likelihood.  At low trial counts:\n",
    "\n",
    "- The likelihood surface is **flat** in some directions, so the optimiser lands at an\n",
    "  arbitrary point in a large region of near-equal likelihood.\n",
    "- **Boundary effects** distort estimates: noise parameters can converge to near-zero\n",
    "  (overfitting a lucky sample) or blow up.\n",
    "- There is **no uncertainty quantification**: you get a point estimate with no error bar,\n",
    "  so you cannot distinguish a well-identified subject from a poorly-identified one.\n",
    "\n",
    "Hierarchical Bayes solves all three: the group prior tilts the likelihood surface toward\n",
    "sensible values, the posterior is a full distribution (not a point), and the amount of\n",
    "shrinkage is automatically calibrated per subject.\n",
    "\n",
    "### Rule of thumb\n",
    "\n",
    "If you have < 200 trials per subject per condition, **always use hierarchical fitting**.\n",
    "If you have < 100, individual fitting is essentially meaningless for models with more\n",
    "than one or two parameters.  bauer makes hierarchical fitting the default for exactly\n",
    "this reason."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "ce5b9486",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-04-10T07:52:03.923959Z",
     "iopub.status.busy": "2026-04-10T07:52:03.923851Z",
     "iopub.status.idle": "2026-04-10T08:13:39.195604Z",
     "shell.execute_reply": "2026-04-10T08:13:39.195028Z"
    }
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Initializing NUTS using jitter+adapt_diag...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Multiprocess sampling (2 chains in 2 jobs)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "NUTS: [n1_evidence_sd_mu_untransformed, n1_evidence_sd_sd, n1_evidence_sd_offset, n2_evidence_sd_mu_untransformed, n2_evidence_sd_sd, n2_evidence_sd_offset]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Sampling 2 chains for 500 tune and 500 draw iterations (1_000 + 1_000 draws total) took 5 seconds.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "We recommend running at least 4 chains for robust computation of convergence diagnostics\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "The rhat statistic is larger than 1.01 for some parameters. This indicates problems during sampling. See https://arxiv.org/abs/1903.08008 for details\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "The effective sample size per chain is smaller than 100 for some parameters.  A higher number is needed for reliable rhat and ess computation. See https://arxiv.org/abs/1903.08008 for details\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Initializing NUTS using jitter+adapt_diag...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Multiprocess sampling (2 chains in 2 jobs)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "NUTS: [n1_evidence_sd_mu_untransformed, n1_evidence_sd_sd, n1_evidence_sd_offset, n2_evidence_sd_mu_untransformed, n2_evidence_sd_sd, n2_evidence_sd_offset]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Sampling 2 chains for 500 tune and 500 draw iterations (1_000 + 1_000 draws total) took 5 seconds.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "We recommend running at least 4 chains for robust computation of convergence diagnostics\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "The rhat statistic is larger than 1.01 for some parameters. This indicates problems during sampling. See https://arxiv.org/abs/1903.08008 for details\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "The effective sample size per chain is smaller than 100 for some parameters.  A higher number is needed for reliable rhat and ess computation. See https://arxiv.org/abs/1903.08008 for details\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Initializing NUTS using jitter+adapt_diag...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Multiprocess sampling (2 chains in 2 jobs)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "NUTS: [n1_evidence_sd_mu_untransformed, n1_evidence_sd_sd, n1_evidence_sd_offset, n2_evidence_sd_mu_untransformed, n2_evidence_sd_sd, n2_evidence_sd_offset]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Sampling 2 chains for 500 tune and 500 draw iterations (1_000 + 1_000 draws total) took 12 seconds.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "There were 31 divergences after tuning. Increase `target_accept` or reparameterize.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "We recommend running at least 4 chains for robust computation of convergence diagnostics\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "The rhat statistic is larger than 1.01 for some parameters. This indicates problems during sampling. See https://arxiv.org/abs/1903.08008 for details\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "The effective sample size per chain is smaller than 100 for some parameters.  A higher number is needed for reliable rhat and ess computation. See https://arxiv.org/abs/1903.08008 for details\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Initializing NUTS using jitter+adapt_diag...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Multiprocess sampling (2 chains in 2 jobs)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "NUTS: [n1_evidence_sd_mu_untransformed, n1_evidence_sd_sd, n1_evidence_sd_offset, n2_evidence_sd_mu_untransformed, n2_evidence_sd_sd, n2_evidence_sd_offset]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Sampling 2 chains for 500 tune and 500 draw iterations (1_000 + 1_000 draws total) took 10 seconds.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "There were 13 divergences after tuning. Increase `target_accept` or reparameterize.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "We recommend running at least 4 chains for robust computation of convergence diagnostics\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "The rhat statistic is larger than 1.01 for some parameters. This indicates problems during sampling. See https://arxiv.org/abs/1903.08008 for details\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "The effective sample size per chain is smaller than 100 for some parameters.  A higher number is needed for reliable rhat and ess computation. See https://arxiv.org/abs/1903.08008 for details\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Initializing NUTS using jitter+adapt_diag...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Multiprocess sampling (2 chains in 2 jobs)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "NUTS: [n1_evidence_sd_mu_untransformed, n1_evidence_sd_sd, n1_evidence_sd_offset, n2_evidence_sd_mu_untransformed, n2_evidence_sd_sd, n2_evidence_sd_offset]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Sampling 2 chains for 500 tune and 500 draw iterations (1_000 + 1_000 draws total) took 29 seconds.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "There were 8 divergences after tuning. Increase `target_accept` or reparameterize.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "We recommend running at least 4 chains for robust computation of convergence diagnostics\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "The rhat statistic is larger than 1.01 for some parameters. This indicates problems during sampling. See https://arxiv.org/abs/1903.08008 for details\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "The effective sample size per chain is smaller than 100 for some parameters.  A higher number is needed for reliable rhat and ess computation. See https://arxiv.org/abs/1903.08008 for details\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Initializing NUTS using jitter+adapt_diag...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Multiprocess sampling (2 chains in 2 jobs)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "NUTS: [n1_evidence_sd_mu_untransformed, n1_evidence_sd_sd, n1_evidence_sd_offset, n2_evidence_sd_mu_untransformed, n2_evidence_sd_sd, n2_evidence_sd_offset]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Sampling 2 chains for 500 tune and 500 draw iterations (1_000 + 1_000 draws total) took 22 seconds.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "There were 309 divergences after tuning. Increase `target_accept` or reparameterize.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "We recommend running at least 4 chains for robust computation of convergence diagnostics\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "The rhat statistic is larger than 1.01 for some parameters. This indicates problems during sampling. See https://arxiv.org/abs/1903.08008 for details\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "The effective sample size per chain is smaller than 100 for some parameters.  A higher number is needed for reliable rhat and ess computation. See https://arxiv.org/abs/1903.08008 for details\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Initializing NUTS using jitter+adapt_diag...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Multiprocess sampling (2 chains in 2 jobs)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "NUTS: [n1_evidence_sd_mu_untransformed, n1_evidence_sd_sd, n1_evidence_sd_offset, n2_evidence_sd_mu_untransformed, n2_evidence_sd_sd, n2_evidence_sd_offset]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Sampling 2 chains for 500 tune and 500 draw iterations (1_000 + 1_000 draws total) took 5 seconds.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "We recommend running at least 4 chains for robust computation of convergence diagnostics\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "The rhat statistic is larger than 1.01 for some parameters. This indicates problems during sampling. See https://arxiv.org/abs/1903.08008 for details\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "The effective sample size per chain is smaller than 100 for some parameters.  A higher number is needed for reliable rhat and ess computation. See https://arxiv.org/abs/1903.08008 for details\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Initializing NUTS using jitter+adapt_diag...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Multiprocess sampling (2 chains in 2 jobs)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "NUTS: [n1_evidence_sd_mu_untransformed, n1_evidence_sd_sd, n1_evidence_sd_offset, n2_evidence_sd_mu_untransformed, n2_evidence_sd_sd, n2_evidence_sd_offset]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Sampling 2 chains for 500 tune and 500 draw iterations (1_000 + 1_000 draws total) took 5 seconds.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "There was 1 divergence after tuning. Increase `target_accept` or reparameterize.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "We recommend running at least 4 chains for robust computation of convergence diagnostics\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "The rhat statistic is larger than 1.01 for some parameters. This indicates problems during sampling. See https://arxiv.org/abs/1903.08008 for details\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Initializing NUTS using jitter+adapt_diag...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Multiprocess sampling (2 chains in 2 jobs)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "NUTS: [n1_evidence_sd_mu_untransformed, n1_evidence_sd_sd, n1_evidence_sd_offset, n2_evidence_sd_mu_untransformed, n2_evidence_sd_sd, n2_evidence_sd_offset]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Sampling 2 chains for 500 tune and 500 draw iterations (1_000 + 1_000 draws total) took 9 seconds.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "There were 5 divergences after tuning. Increase `target_accept` or reparameterize.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "We recommend running at least 4 chains for robust computation of convergence diagnostics\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "The rhat statistic is larger than 1.01 for some parameters. This indicates problems during sampling. See https://arxiv.org/abs/1903.08008 for details\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Initializing NUTS using jitter+adapt_diag...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Multiprocess sampling (2 chains in 2 jobs)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "NUTS: [n1_evidence_sd_mu_untransformed, n1_evidence_sd_sd, n1_evidence_sd_offset, n2_evidence_sd_mu_untransformed, n2_evidence_sd_sd, n2_evidence_sd_offset]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Sampling 2 chains for 500 tune and 500 draw iterations (1_000 + 1_000 draws total) took 10 seconds.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "There were 8 divergences after tuning. Increase `target_accept` or reparameterize.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "We recommend running at least 4 chains for robust computation of convergence diagnostics\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "The rhat statistic is larger than 1.01 for some parameters. This indicates problems during sampling. See https://arxiv.org/abs/1903.08008 for details\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "The effective sample size per chain is smaller than 100 for some parameters.  A higher number is needed for reliable rhat and ess computation. See https://arxiv.org/abs/1903.08008 for details\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Initializing NUTS using jitter+adapt_diag...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Multiprocess sampling (2 chains in 2 jobs)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "NUTS: [n1_evidence_sd_mu_untransformed, n1_evidence_sd_sd, n1_evidence_sd_offset, n2_evidence_sd_mu_untransformed, n2_evidence_sd_sd, n2_evidence_sd_offset]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Sampling 2 chains for 500 tune and 500 draw iterations (1_000 + 1_000 draws total) took 29 seconds.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "There were 45 divergences after tuning. Increase `target_accept` or reparameterize.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "We recommend running at least 4 chains for robust computation of convergence diagnostics\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "The rhat statistic is larger than 1.01 for some parameters. This indicates problems during sampling. See https://arxiv.org/abs/1903.08008 for details\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "The effective sample size per chain is smaller than 100 for some parameters.  A higher number is needed for reliable rhat and ess computation. See https://arxiv.org/abs/1903.08008 for details\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Initializing NUTS using jitter+adapt_diag...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Multiprocess sampling (2 chains in 2 jobs)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "NUTS: [n1_evidence_sd_mu_untransformed, n1_evidence_sd_sd, n1_evidence_sd_offset, n2_evidence_sd_mu_untransformed, n2_evidence_sd_sd, n2_evidence_sd_offset]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Sampling 2 chains for 500 tune and 500 draw iterations (1_000 + 1_000 draws total) took 27 seconds.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "There were 53 divergences after tuning. Increase `target_accept` or reparameterize.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "We recommend running at least 4 chains for robust computation of convergence diagnostics\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "The rhat statistic is larger than 1.01 for some parameters. This indicates problems during sampling. See https://arxiv.org/abs/1903.08008 for details\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "The effective sample size per chain is smaller than 100 for some parameters.  A higher number is needed for reliable rhat and ess computation. See https://arxiv.org/abs/1903.08008 for details\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Initializing NUTS using jitter+adapt_diag...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Multiprocess sampling (2 chains in 2 jobs)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "NUTS: [n1_evidence_sd_mu_untransformed, n1_evidence_sd_sd, n1_evidence_sd_offset, n2_evidence_sd_mu_untransformed, n2_evidence_sd_sd, n2_evidence_sd_offset]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Sampling 2 chains for 500 tune and 500 draw iterations (1_000 + 1_000 draws total) took 5 seconds.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "There was 1 divergence after tuning. Increase `target_accept` or reparameterize.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "We recommend running at least 4 chains for robust computation of convergence diagnostics\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "The rhat statistic is larger than 1.01 for some parameters. This indicates problems during sampling. See https://arxiv.org/abs/1903.08008 for details\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Initializing NUTS using jitter+adapt_diag...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Multiprocess sampling (2 chains in 2 jobs)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "NUTS: [n1_evidence_sd_mu_untransformed, n1_evidence_sd_sd, n1_evidence_sd_offset, n2_evidence_sd_mu_untransformed, n2_evidence_sd_sd, n2_evidence_sd_offset]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Sampling 2 chains for 500 tune and 500 draw iterations (1_000 + 1_000 draws total) took 5 seconds.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "There was 1 divergence after tuning. Increase `target_accept` or reparameterize.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "We recommend running at least 4 chains for robust computation of convergence diagnostics\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "The rhat statistic is larger than 1.01 for some parameters. This indicates problems during sampling. See https://arxiv.org/abs/1903.08008 for details\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "The effective sample size per chain is smaller than 100 for some parameters.  A higher number is needed for reliable rhat and ess computation. See https://arxiv.org/abs/1903.08008 for details\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Initializing NUTS using jitter+adapt_diag...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Multiprocess sampling (2 chains in 2 jobs)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "NUTS: [n1_evidence_sd_mu_untransformed, n1_evidence_sd_sd, n1_evidence_sd_offset, n2_evidence_sd_mu_untransformed, n2_evidence_sd_sd, n2_evidence_sd_offset]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Sampling 2 chains for 500 tune and 500 draw iterations (1_000 + 1_000 draws total) took 10 seconds.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "There were 24 divergences after tuning. Increase `target_accept` or reparameterize.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "We recommend running at least 4 chains for robust computation of convergence diagnostics\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "The rhat statistic is larger than 1.01 for some parameters. This indicates problems during sampling. See https://arxiv.org/abs/1903.08008 for details\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "The effective sample size per chain is smaller than 100 for some parameters.  A higher number is needed for reliable rhat and ess computation. See https://arxiv.org/abs/1903.08008 for details\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Initializing NUTS using jitter+adapt_diag...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Multiprocess sampling (2 chains in 2 jobs)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "NUTS: [n1_evidence_sd_mu_untransformed, n1_evidence_sd_sd, n1_evidence_sd_offset, n2_evidence_sd_mu_untransformed, n2_evidence_sd_sd, n2_evidence_sd_offset]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Sampling 2 chains for 500 tune and 500 draw iterations (1_000 + 1_000 draws total) took 10 seconds.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "There were 10 divergences after tuning. Increase `target_accept` or reparameterize.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "We recommend running at least 4 chains for robust computation of convergence diagnostics\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "The rhat statistic is larger than 1.01 for some parameters. This indicates problems during sampling. See https://arxiv.org/abs/1903.08008 for details\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "The effective sample size per chain is smaller than 100 for some parameters.  A higher number is needed for reliable rhat and ess computation. See https://arxiv.org/abs/1903.08008 for details\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Initializing NUTS using jitter+adapt_diag...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Multiprocess sampling (2 chains in 2 jobs)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "NUTS: [n1_evidence_sd_mu_untransformed, n1_evidence_sd_sd, n1_evidence_sd_offset, n2_evidence_sd_mu_untransformed, n2_evidence_sd_sd, n2_evidence_sd_offset]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Sampling 2 chains for 500 tune and 500 draw iterations (1_000 + 1_000 draws total) took 33 seconds.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "There were 6 divergences after tuning. Increase `target_accept` or reparameterize.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "We recommend running at least 4 chains for robust computation of convergence diagnostics\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "The rhat statistic is larger than 1.01 for some parameters. This indicates problems during sampling. See https://arxiv.org/abs/1903.08008 for details\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "The effective sample size per chain is smaller than 100 for some parameters.  A higher number is needed for reliable rhat and ess computation. See https://arxiv.org/abs/1903.08008 for details\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Initializing NUTS using jitter+adapt_diag...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Multiprocess sampling (2 chains in 2 jobs)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "NUTS: [n1_evidence_sd_mu_untransformed, n1_evidence_sd_sd, n1_evidence_sd_offset, n2_evidence_sd_mu_untransformed, n2_evidence_sd_sd, n2_evidence_sd_offset]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Sampling 2 chains for 500 tune and 500 draw iterations (1_000 + 1_000 draws total) took 24 seconds.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "There were 7 divergences after tuning. Increase `target_accept` or reparameterize.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "We recommend running at least 4 chains for robust computation of convergence diagnostics\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "The rhat statistic is larger than 1.01 for some parameters. This indicates problems during sampling. See https://arxiv.org/abs/1903.08008 for details\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "The effective sample size per chain is smaller than 100 for some parameters.  A higher number is needed for reliable rhat and ess computation. See https://arxiv.org/abs/1903.08008 for details\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1050x700 with 6 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1050x700 with 6 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1050x700 with 6 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# ── Scatter: half A vs half B at every trial count, all 3 params ─────────────\n",
    "half_a, half_b = split_data(df_mag, seed=0)\n",
    "scatter_colors = {'MLE': '#d73027', 'Hierarchical Bayes': '#4393c3'}\n",
    "\n",
    "for param, param_label in param_labels.items():\n",
    "    n_k = len(trial_counts)\n",
    "    fig, axes = plt.subplots(2, n_k, figsize=(3.5 * n_k, 7), sharey='row', sharex='row')\n",
    "\n",
    "    for col, k in enumerate(trial_counts):\n",
    "        a_sub = subsample(half_a, k)\n",
    "        b_sub = subsample(half_b, k)\n",
    "        ests = {}\n",
    "        for name, fn in methods.items():\n",
    "            ea, eb = fn(a_sub), fn(b_sub)\n",
    "            ea['total_sd'] = ea['n1_evidence_sd'] + ea['n2_evidence_sd']\n",
    "            eb['total_sd'] = eb['n1_evidence_sd'] + eb['n2_evidence_sd']\n",
    "            ests[name] = (ea, eb)\n",
    "        for row, method in enumerate(['MLE', 'Hierarchical Bayes']):\n",
    "            ax = axes[row, col]\n",
    "            est_a, est_b = ests[method]\n",
    "            rho_p, _ = pearsonr(est_a[param], est_b[param])\n",
    "            ax.scatter(est_a[param], est_b[param], s=25, alpha=.6,\n",
    "                       color=scatter_colors[method])\n",
    "            lims = [0, max(est_a[param].max(), est_b[param].max()) * 1.1]\n",
    "            ax.plot(lims, lims, '--', color='gray', lw=1)\n",
    "            ax.set_xlim(lims); ax.set_ylim(lims)\n",
    "            ax.set_title(f'{method}, k={k}  (R\\u00b2={rho_p**2:.2f})', fontsize=9)\n",
    "            if col == 0:\n",
    "                ax.set_ylabel(f'Half B')\n",
    "            if row == 1:\n",
    "                ax.set_xlabel(f'Half A')\n",
    "            sns.despine(ax=ax)\n",
    "\n",
    "    plt.suptitle(f'Split-half scatter: {param_label}  (MLE top, Hierarchical bottom)',\n",
    "                 fontsize=12, y=1.01)\n",
    "    plt.tight_layout()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9ab4f4f9",
   "metadata": {},
   "source": [
    "## Take-aways\n",
    "\n",
    "- **Never use individual MLE/MAP for cognitive models at typical psychophysics trial\n",
    "  counts.**  The estimates are unreliable and any downstream correlation (e.g. with\n",
    "  neural data or clinical scores) will be attenuated toward zero.\n",
    "- **Hierarchical Bayes is not just \"nicer\" — it is a prerequisite for valid individual-\n",
    "  difference analyses** at the trial counts we actually work with.\n",
    "- bauer defaults to hierarchical fitting (`hierarchical=True`) for exactly this reason.\n",
    "  Individual MAP (`model.fit_map()`) is available for quick sanity checks but should not\n",
    "  be used for final inference.\n",
    "- These results generalise beyond magnitude comparison: the noisier your model and the\n",
    "  fewer your trials, the bigger the hierarchical advantage."
   ]
  }
 ],
 "metadata": {
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.11.13"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}