{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# EP-M03 Companion Notebook\n",
    "\n",
    "## Deciding When Evidence Is Incomplete\n",
    "\n",
    "**Asset:** EP-M03-A08  \n",
    "**Release:** v0.2  \n",
    "**Status:** Public learning companion\n",
    "\n",
    "This self-contained executive lab uses synthetic instructional data. It shows how evidence updates beliefs, how asymmetric losses determine action, and how accountable leaders test whether a recommendation is robust. It does not grant decision authority.\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Learning and decision contract\n",
    "\n",
    "By completing the notebook, you will be able to:\n",
    "\n",
    "1. document a prior and its provenance;\n",
    "2. update beliefs with one or several evidence items;\n",
    "3. calculate posterior expected loss and identify the Bayes action;\n",
    "4. test sensitivity to priors, likelihoods, losses and missing states;\n",
    "5. distinguish perfect-information value from the value of an imperfect test;\n",
    "6. expose distributional effects hidden by aggregate loss; and\n",
    "7. prepare an executive record supporting **ACT**, **PILOT**, **WAIT AND LEARN**, or **ESCALATE**.\n",
    "\n",
    "The canonical worked case is in **Agriculture**. Transfer cases cover **Education, Health and Habitat**.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Python 3.12.14\n",
      "NumPy 2.3.5 | pandas 2.2.3\n",
      "Deterministic seed: 42\n"
     ]
    }
   ],
   "source": [
    "import sys\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "from io import StringIO\n",
    "\n",
    "RANDOM_SEED = 42\n",
    "rng = np.random.default_rng(RANDOM_SEED)\n",
    "np.set_printoptions(precision=4, suppress=True)\n",
    "pd.set_option(\"display.max_columns\", 20)\n",
    "\n",
    "print(f\"Python {sys.version.split()[0]}\")\n",
    "print(f\"NumPy {np.__version__} | pandas {pd.__version__}\")\n",
    "print(f\"Deterministic seed: {RANDOM_SEED}\")\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 1 Transparent decision functions\n",
    "\n",
    "Every function below is visible and testable. Probabilities must be finite and non-negative. A loss matrix has one row per action and one column per uncertain state. Lower loss is better.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "def probability_vector(values, name=\"probabilities\"):\n",
    "    values = np.asarray(values, dtype=float)\n",
    "    if values.ndim != 1 or values.size < 2:\n",
    "        raise ValueError(f\"{name} must contain at least two states.\")\n",
    "    if not np.all(np.isfinite(values)) or np.any(values < 0):\n",
    "        raise ValueError(f\"{name} must be finite and non-negative.\")\n",
    "    total = values.sum()\n",
    "    if total <= 0:\n",
    "        raise ValueError(f\"{name} must have a positive sum.\")\n",
    "    return values / total\n",
    "\n",
    "def posterior(prior, likelihood):\n",
    "    prior = probability_vector(prior, \"prior\")\n",
    "    likelihood = np.asarray(likelihood, dtype=float)\n",
    "    if likelihood.shape != prior.shape:\n",
    "        raise ValueError(\"likelihood must have the same shape as prior.\")\n",
    "    if not np.all(np.isfinite(likelihood)) or np.any(likelihood < 0):\n",
    "        raise ValueError(\"likelihood must be finite and non-negative.\")\n",
    "    evidence_probability = float(prior @ likelihood)\n",
    "    if evidence_probability <= 0:\n",
    "        raise ValueError(\"observed evidence has zero probability under the model.\")\n",
    "    return prior * likelihood / evidence_probability\n",
    "\n",
    "def expected_losses(loss_matrix, beliefs):\n",
    "    beliefs = probability_vector(beliefs, \"beliefs\")\n",
    "    losses = np.asarray(loss_matrix, dtype=float)\n",
    "    if losses.ndim != 2 or losses.shape[1] != beliefs.size:\n",
    "        raise ValueError(\"loss_matrix must have one column per state.\")\n",
    "    if not np.all(np.isfinite(losses)):\n",
    "        raise ValueError(\"loss_matrix must contain finite values.\")\n",
    "    return losses @ beliefs\n",
    "\n",
    "def choose_action(loss_matrix, beliefs, action_labels):\n",
    "    values = expected_losses(loss_matrix, beliefs)\n",
    "    if len(action_labels) != values.size:\n",
    "        raise ValueError(\"action_labels must match loss-matrix rows.\")\n",
    "    index = int(np.argmin(values))\n",
    "    return {\"action\": action_labels[index], \"index\": index, \"expected_losses\": values}\n",
    "\n",
    "def evpi_from_losses(loss_matrix, beliefs):\n",
    "    beliefs = probability_vector(beliefs, \"beliefs\")\n",
    "    losses = np.asarray(loss_matrix, dtype=float)\n",
    "    current_best = float(np.min(losses @ beliefs))\n",
    "    loss_with_perfect_information = float(np.sum(beliefs * np.min(losses, axis=0)))\n",
    "    return current_best - loss_with_perfect_information\n",
    "\n",
    "def expected_value_sample_information(loss_matrix, beliefs, outcome_likelihoods):\n",
    "    beliefs = probability_vector(beliefs, \"beliefs\")\n",
    "    likelihoods = np.asarray(outcome_likelihoods, dtype=float)\n",
    "    if likelihoods.ndim != 2 or likelihoods.shape[1] != beliefs.size:\n",
    "        raise ValueError(\"outcome_likelihoods must have one column per state.\")\n",
    "    if not np.allclose(likelihoods.sum(axis=0), 1.0):\n",
    "        raise ValueError(\"outcome likelihoods must sum to one within each state.\")\n",
    "    current_best = float(np.min(np.asarray(loss_matrix) @ beliefs))\n",
    "    expected_loss_after_sample = 0.0\n",
    "    details = []\n",
    "    for outcome_index, likelihood in enumerate(likelihoods):\n",
    "        outcome_probability = float(beliefs @ likelihood)\n",
    "        updated = posterior(beliefs, likelihood)\n",
    "        decision = choose_action(loss_matrix, updated, list(range(np.asarray(loss_matrix).shape[0])))\n",
    "        best_loss = float(np.min(decision[\"expected_losses\"]))\n",
    "        expected_loss_after_sample += outcome_probability * best_loss\n",
    "        details.append((outcome_index, outcome_probability, updated, best_loss, decision[\"index\"]))\n",
    "    return current_best - expected_loss_after_sample, details\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "PASS: core probability and validation functions.\n"
     ]
    }
   ],
   "source": [
    "# Immediate unit checks for the reusable functions\n",
    "assert np.allclose(probability_vector([2, 3]), [0.4, 0.6])\n",
    "assert np.allclose(posterior([0.6, 0.4], [0.3, 0.8]), [0.36, 0.64])\n",
    "try:\n",
    "    posterior([0, 0], [0.3, 0.8])\n",
    "except ValueError:\n",
    "    pass\n",
    "else:\n",
    "    raise AssertionError(\"A zero-sum prior must fail.\")\n",
    "print(\"PASS: core probability and validation functions.\")\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 2 Canonical Agriculture case\n",
    "\n",
    "A programme authority must decide whether to deploy early crop-protection support. Conditions are either **stable** or **adverse**. New field evidence is more likely under adverse conditions. Losses are synthetic decision units combining operational, livelihood and delay consequences.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  Element                                  Declared value\n",
      " Decision            Deploy early crop-protection support\n",
      "Authority               Programme authorization committee\n",
      "  Horizon                        Current production cycle\n",
      "   States                    Stable or adverse conditions\n",
      "  Actions                   Do not intervene or intervene\n",
      " Evidence Field signal consistent with adverse conditions\n"
     ]
    }
   ],
   "source": [
    "states = [\"Stable conditions\", \"Adverse conditions\"]\n",
    "actions = [\"Do not intervene\", \"Intervene\"]\n",
    "prior = np.array([0.60, 0.40])\n",
    "likelihood_of_field_evidence = np.array([0.30, 0.80])\n",
    "loss_matrix = np.array([[0.0, 20.0], [8.0, 0.0]])\n",
    "\n",
    "case_register = pd.DataFrame({\n",
    "    \"Element\": [\"Decision\", \"Authority\", \"Horizon\", \"States\", \"Actions\", \"Evidence\"],\n",
    "    \"Declared value\": [\n",
    "        \"Deploy early crop-protection support\",\n",
    "        \"Programme authorization committee\",\n",
    "        \"Current production cycle\",\n",
    "        \"Stable or adverse conditions\",\n",
    "        \"Do not intervene or intervene\",\n",
    "        \"Field signal consistent with adverse conditions\",\n",
    "    ],\n",
    "})\n",
    "print(case_register.to_string(index=False))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "             State  Prior  Likelihood of evidence  Joint weight  Posterior\n",
      " Stable conditions    0.6                     0.3          0.18       0.36\n",
      "Adverse conditions    0.4                     0.8          0.32       0.64\n",
      "\n",
      "                  Stable conditions  Adverse conditions  Posterior expected loss\n",
      "Do not intervene                0.0                20.0                    12.80\n",
      "Intervene                       8.0                 0.0                     2.88\n",
      "\n",
      "Analytical preference: Intervene\n",
      "EVPI upper bound: 2.880 loss units\n"
     ]
    }
   ],
   "source": [
    "post = posterior(prior, likelihood_of_field_evidence)\n",
    "decision = choose_action(loss_matrix, post, actions)\n",
    "evpi = evpi_from_losses(loss_matrix, post)\n",
    "\n",
    "update_table = pd.DataFrame({\n",
    "    \"State\": states,\n",
    "    \"Prior\": prior,\n",
    "    \"Likelihood of evidence\": likelihood_of_field_evidence,\n",
    "    \"Joint weight\": prior * likelihood_of_field_evidence,\n",
    "    \"Posterior\": post,\n",
    "})\n",
    "loss_table = pd.DataFrame(loss_matrix, index=actions, columns=states)\n",
    "loss_table[\"Posterior expected loss\"] = decision[\"expected_losses\"]\n",
    "\n",
    "print(update_table.round(4).to_string(index=False))\n",
    "print()\n",
    "print(loss_table.round(4).to_string())\n",
    "print(f\"\\nAnalytical preference: {decision['action']}\")\n",
    "print(f\"EVPI upper bound: {evpi:.3f} loss units\")\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Interpretation\n",
    "\n",
    "The evidence moves the adverse-state belief from 0.40 to 0.64. Intervention is preferred because its expected loss is 2.88 rather than 12.80. The recommendation depends on both the posterior and the loss matrix. A posterior probability never authorizes action by itself.\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 3 Sequential evidence\n",
    "\n",
    "Evidence often arrives in stages. Updating should be ordered, documented and checked for dependence. The example treats the two evidence items as conditionally independent given the state; that assumption must be challenged in a real decision.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                        Evidence  Prior adverse  Posterior adverse Preferred action  Minimum expected loss\n",
      "                    Field signal           0.40             0.6400        Intervene                 2.8800\n",
      "Independent remote-sensing alert           0.64             0.7692        Intervene                 1.8462\n"
     ]
    }
   ],
   "source": [
    "evidence_sequence = [\n",
    "    (\"Field signal\", np.array([0.30, 0.80])),\n",
    "    (\"Independent remote-sensing alert\", np.array([0.40, 0.75])),\n",
    "]\n",
    "\n",
    "current = prior.copy()\n",
    "sequential_rows = []\n",
    "for evidence_name, likelihood in evidence_sequence:\n",
    "    before = current.copy()\n",
    "    current = posterior(current, likelihood)\n",
    "    selected = choose_action(loss_matrix, current, actions)\n",
    "    sequential_rows.append({\n",
    "        \"Evidence\": evidence_name,\n",
    "        \"Prior adverse\": before[1],\n",
    "        \"Posterior adverse\": current[1],\n",
    "        \"Preferred action\": selected[\"action\"],\n",
    "        \"Minimum expected loss\": selected[\"expected_losses\"].min(),\n",
    "    })\n",
    "\n",
    "sequential_summary = pd.DataFrame(sequential_rows)\n",
    "print(sequential_summary.round(4).to_string(index=False))\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 4 Prior provenance and disagreement\n",
    "\n",
    "A prior can come from historical frequency, expert elicitation, a comparable programme or a formal model. Rather than hiding disagreement, compare credible priors and show whether they alter the action.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              Prior source  Prior adverse  Posterior adverse  Loss no intervention  Loss intervention Preference\n",
      "      Historical base rate           0.25              0.471                 9.412              4.235  Intervene\n",
      "Current programme estimate           0.40              0.640                12.800              2.880  Intervene\n",
      " Precautionary expert view           0.60              0.800                16.000              1.600  Intervene\n"
     ]
    }
   ],
   "source": [
    "prior_scenarios = {\n",
    "    \"Historical base rate\": np.array([0.75, 0.25]),\n",
    "    \"Current programme estimate\": np.array([0.60, 0.40]),\n",
    "    \"Precautionary expert view\": np.array([0.40, 0.60]),\n",
    "}\n",
    "\n",
    "prior_rows = []\n",
    "for source, candidate_prior in prior_scenarios.items():\n",
    "    candidate_post = posterior(candidate_prior, likelihood_of_field_evidence)\n",
    "    candidate_decision = choose_action(loss_matrix, candidate_post, actions)\n",
    "    prior_rows.append({\n",
    "        \"Prior source\": source,\n",
    "        \"Prior adverse\": candidate_prior[1],\n",
    "        \"Posterior adverse\": candidate_post[1],\n",
    "        \"Loss no intervention\": candidate_decision[\"expected_losses\"][0],\n",
    "        \"Loss intervention\": candidate_decision[\"expected_losses\"][1],\n",
    "        \"Preference\": candidate_decision[\"action\"],\n",
    "    })\n",
    "prior_comparison = pd.DataFrame(prior_rows)\n",
    "print(prior_comparison.round(3).to_string(index=False))\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 5 Likelihood uncertainty\n",
    "\n",
    "The evidence model may itself be uncertain. We sample plausible likelihoods around the declared values and observe how often each action is preferred. This is a sensitivity exercise, not a claim that the sampled distribution is objectively correct.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Preference\n",
      "Intervene    1.0\n",
      "Name: Share, dtype: float64\n",
      "count    5000.0000\n",
      "mean        0.6400\n",
      "std         0.0370\n",
      "min         0.5127\n",
      "5%          0.5801\n",
      "50%         0.6394\n",
      "95%         0.7023\n",
      "max         0.7657\n",
      "Name: Posterior adverse, dtype: float64\n"
     ]
    },
    {
     "data": {
      "image/png": "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"
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "draws = 5000\n",
    "likelihood_stable_draws = rng.beta(30, 70, size=draws)\n",
    "likelihood_adverse_draws = rng.beta(80, 20, size=draws)\n",
    "\n",
    "likelihood_rows = []\n",
    "for stable_like, adverse_like in zip(likelihood_stable_draws, likelihood_adverse_draws):\n",
    "    sampled_post = posterior(prior, [stable_like, adverse_like])\n",
    "    sampled_decision = choose_action(loss_matrix, sampled_post, actions)\n",
    "    likelihood_rows.append((sampled_post[1], sampled_decision[\"action\"]))\n",
    "likelihood_sensitivity = pd.DataFrame(likelihood_rows, columns=[\"Posterior adverse\", \"Preference\"])\n",
    "\n",
    "print(likelihood_sensitivity[\"Preference\"].value_counts(normalize=True).rename(\"Share\").round(4))\n",
    "print(likelihood_sensitivity[\"Posterior adverse\"].describe(percentiles=[0.05, 0.5, 0.95]).round(4))\n",
    "\n",
    "fig, ax = plt.subplots(figsize=(8, 4.2))\n",
    "ax.hist(likelihood_sensitivity[\"Posterior adverse\"], bins=35, color=\"#2B6F77\", alpha=0.85)\n",
    "ax.axvline(post[1], color=\"#9C6B18\", linestyle=\"--\", label=\"Canonical posterior\")\n",
    "ax.set(title=\"Posterior uncertainty from likelihood uncertainty\", xlabel=\"Posterior probability of adverse conditions\", ylabel=\"Simulation count\")\n",
    "ax.legend()\n",
    "plt.tight_layout()\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 6 Prior and loss sensitivity\n",
    "\n",
    "The recommendation can reverse because beliefs change or because leaders assign different consequences to acting unnecessarily and failing to act. The grid below makes both sources of fragility visible.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "With the canonical losses, intervention first becomes preferred near prior adverse = 0.140.\n"
     ]
    },
    {
     "data": {
      "image/png": "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"
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "prior_adverse_grid = np.linspace(0.02, 0.98, 97)\n",
    "unnecessary_intervention_losses = np.linspace(2, 18, 65)\n",
    "preference_grid = np.zeros((len(unnecessary_intervention_losses), len(prior_adverse_grid)))\n",
    "\n",
    "for i, unnecessary_loss in enumerate(unnecessary_intervention_losses):\n",
    "    candidate_losses = np.array([[0.0, 20.0], [unnecessary_loss, 0.0]])\n",
    "    for j, prior_adverse in enumerate(prior_adverse_grid):\n",
    "        candidate_post = posterior([1 - prior_adverse, prior_adverse], likelihood_of_field_evidence)\n",
    "        preference_grid[i, j] = choose_action(candidate_losses, candidate_post, actions)[\"index\"]\n",
    "\n",
    "canonical_line = []\n",
    "for prior_adverse in prior_adverse_grid:\n",
    "    candidate_post = posterior([1-prior_adverse, prior_adverse], likelihood_of_field_evidence)\n",
    "    canonical_line.append(choose_action(loss_matrix, candidate_post, actions)[\"index\"])\n",
    "reversal_prior = float(prior_adverse_grid[np.where(np.array(canonical_line) == 1)[0][0]])\n",
    "print(f\"With the canonical losses, intervention first becomes preferred near prior adverse = {reversal_prior:.3f}.\")\n",
    "\n",
    "fig, ax = plt.subplots(figsize=(8.5, 4.8))\n",
    "image = ax.imshow(preference_grid, origin=\"lower\", aspect=\"auto\", extent=[prior_adverse_grid.min(), prior_adverse_grid.max(), unnecessary_intervention_losses.min(), unnecessary_intervention_losses.max()], cmap=\"cividis\")\n",
    "ax.scatter([prior[1]], [loss_matrix[1,0]], color=\"red\", s=55, label=\"Canonical assumption\")\n",
    "ax.set(title=\"Decision regions under prior and loss uncertainty\", xlabel=\"Prior probability of adverse conditions\", ylabel=\"Loss of unnecessary intervention\")\n",
    "ax.legend(loc=\"upper left\")\n",
    "plt.colorbar(image, ax=ax, ticks=[0,1], label=\"0 = Do not intervene, 1 = Intervene\")\n",
    "plt.tight_layout()\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 7 Value of information\n",
    "\n",
    "EVPI assumes uncertainty can be removed perfectly. A real test is imperfect, so its expected value of sample information (EVSI) is normally lower. Evidence should be acquired only when expected decision improvement exceeds acquisition cost, delay and implementation burden.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test outcome  Outcome probability  Posterior adverse Preferred action  Minimum expected loss\n",
      "    Positive                0.552             0.8696        Intervene                 1.0435\n",
      "    Negative                0.448             0.3571        Intervene                 5.1429\n",
      "\n",
      "EVPI: 2.880\n",
      "EVSI for the imperfect test: 0.000\n"
     ]
    }
   ],
   "source": [
    "# Prospective test outcomes: positive and negative. Columns are stable and adverse states.\n",
    "prospective_test = np.array([\n",
    "    [0.20, 0.75],  # positive result\n",
    "    [0.80, 0.25],  # negative result\n",
    "])\n",
    "evsi, sample_details = expected_value_sample_information(loss_matrix, post, prospective_test)\n",
    "outcome_names = [\"Positive\", \"Negative\"]\n",
    "sample_rows = []\n",
    "for outcome_index, outcome_probability, updated, best_loss, action_index in sample_details:\n",
    "    sample_rows.append({\n",
    "        \"Test outcome\": outcome_names[outcome_index],\n",
    "        \"Outcome probability\": outcome_probability,\n",
    "        \"Posterior adverse\": updated[1],\n",
    "        \"Preferred action\": actions[action_index],\n",
    "        \"Minimum expected loss\": best_loss,\n",
    "    })\n",
    "sample_summary = pd.DataFrame(sample_rows)\n",
    "print(sample_summary.round(4).to_string(index=False))\n",
    "print(f\"\\nEVPI: {evpi:.3f}\")\n",
    "print(f\"EVSI for the imperfect test: {evsi:.3f}\")\n",
    "assert evsi <= evpi + 1e-12\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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"
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "test_costs = np.linspace(0, max(evpi * 1.15, 0.1), 100)\n",
    "net_values = evsi - test_costs\n",
    "\n",
    "fig, ax = plt.subplots(figsize=(8, 4.2))\n",
    "ax.plot(test_costs, net_values, color=\"#17365D\")\n",
    "ax.axhline(0, color=\"black\", linewidth=1)\n",
    "ax.axvline(evsi, color=\"#9C6B18\", linestyle=\"--\", label=f\"Break-even cost = {evsi:.3f}\")\n",
    "ax.set(title=\"Net value of the prospective test\", xlabel=\"Test cost including delay\", ylabel=\"EVSI minus cost\")\n",
    "ax.grid(alpha=0.25)\n",
    "ax.legend()\n",
    "plt.tight_layout()\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 8 Missing-state and model-boundary test\n",
    "\n",
    "A two-state model can omit a severe but rare condition. Adding a plausible third state tests whether the original recommendation and available actions remain adequate.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                      State  Posterior\n",
      "                     Stable     0.3273\n",
      "                    Adverse     0.5818\n",
      "Severe unmodelled condition     0.0909\n",
      "\n",
      "                      Stable  Adverse  Severe unmodelled condition  Posterior expected loss\n",
      "Do not intervene         0.0     20.0                         60.0                   17.091\n",
      "Intervene                8.0      0.0                         30.0                    5.345\n",
      "Emergency escalation    18.0     10.0                          2.0                   11.891\n",
      "\n",
      "Preference after adding the missing state: Intervene\n"
     ]
    }
   ],
   "source": [
    "expanded_states = [\"Stable\", \"Adverse\", \"Severe unmodelled condition\"]\n",
    "expanded_actions = [\"Do not intervene\", \"Intervene\", \"Emergency escalation\"]\n",
    "expanded_prior = np.array([0.57, 0.38, 0.05])\n",
    "expanded_likelihood = np.array([0.30, 0.80, 0.95])\n",
    "expanded_losses = np.array([\n",
    "    [0.0, 20.0, 60.0],\n",
    "    [8.0, 0.0, 30.0],\n",
    "    [18.0, 10.0, 2.0],\n",
    "])\n",
    "expanded_post = posterior(expanded_prior, expanded_likelihood)\n",
    "expanded_decision = choose_action(expanded_losses, expanded_post, expanded_actions)\n",
    "expanded_table = pd.DataFrame(expanded_losses, index=expanded_actions, columns=expanded_states)\n",
    "expanded_table[\"Posterior expected loss\"] = expanded_decision[\"expected_losses\"]\n",
    "print(pd.DataFrame({\"State\": expanded_states, \"Posterior\": expanded_post}).round(4).to_string(index=False))\n",
    "print()\n",
    "print(expanded_table.round(3).to_string())\n",
    "print(f\"\\nPreference after adding the missing state: {expanded_decision['action']}\")\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 9 Distributional visibility\n",
    "\n",
    "An aggregate recommendation can conceal who bears the losses. The table separates consequences for smallholder households, programme operations and the public budget. These values are illustrative and should be elicited from affected groups and accountable owners.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "        Affected group  Loss no intervention  Loss intervention Lower-loss action\n",
      "Smallholder households                 17.92               1.80         Intervene\n",
      "  Programme operations                  6.40               3.24         Intervene\n",
      "         Public budget                  8.96               4.32         Intervene\n",
      "\n",
      "Decision rule: aggregate efficiency does not erase distributional or rights-based constraints.\n"
     ]
    }
   ],
   "source": [
    "group_losses = {\n",
    "    \"Smallholder households\": np.array([[0.0, 28.0], [5.0, 0.0]]),\n",
    "    \"Programme operations\": np.array([[0.0, 10.0], [9.0, 0.0]]),\n",
    "    \"Public budget\": np.array([[0.0, 14.0], [12.0, 0.0]]),\n",
    "}\n",
    "distribution_rows = []\n",
    "for group, matrix in group_losses.items():\n",
    "    values = expected_losses(matrix, post)\n",
    "    distribution_rows.append({\"Affected group\": group, \"Loss no intervention\": values[0], \"Loss intervention\": values[1], \"Lower-loss action\": actions[int(np.argmin(values))]})\n",
    "distribution_table = pd.DataFrame(distribution_rows)\n",
    "print(distribution_table.round(3).to_string(index=False))\n",
    "print(\"\\nDecision rule: aggregate efficiency does not erase distributional or rights-based constraints.\")\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 10 Transfer across priority sectors\n",
    "\n",
    "The calculation is reusable, but sector owners must define credible states, evidence and losses. The examples below are synthetic prompts for executive discussion.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "     Sector                         Decision  Prior adverse  Posterior adverse  Loss Wait  Loss Act Preference  EVPI\n",
      "Agriculture   Deploy crop-protection support           0.40              0.640     12.800     2.880        Act 2.880\n",
      "  Education Deploy targeted learning support           0.30              0.562     10.125     2.625        Act 2.625\n",
      "     Health    Escalate diagnostic follow-up           0.20              0.586     17.586     2.897        Act 2.897\n",
      "    Habitat  Authorize structural inspection           0.15              0.414     10.345     2.931        Act 2.931\n"
     ]
    }
   ],
   "source": [
    "sector_cases = [\n",
    "    (\"Agriculture\", \"Deploy crop-protection support\", 0.40, 0.30, 0.80, 20, 8),\n",
    "    (\"Education\", \"Deploy targeted learning support\", 0.30, 0.25, 0.75, 18, 6),\n",
    "    (\"Health\", \"Escalate diagnostic follow-up\", 0.20, 0.15, 0.85, 30, 7),\n",
    "    (\"Habitat\", \"Authorize structural inspection\", 0.15, 0.20, 0.80, 25, 5),\n",
    "]\n",
    "sector_rows = []\n",
    "for sector, question, prior_adverse, like_stable, like_adverse, loss_wait, loss_act in sector_cases:\n",
    "    sector_post = posterior([1-prior_adverse, prior_adverse], [like_stable, like_adverse])\n",
    "    sector_losses = np.array([[0.0, loss_wait], [loss_act, 0.0]])\n",
    "    sector_choice = choose_action(sector_losses, sector_post, [\"Wait\", \"Act\"])\n",
    "    sector_rows.append({\n",
    "        \"Sector\": sector, \"Decision\": question, \"Prior adverse\": prior_adverse,\n",
    "        \"Posterior adverse\": sector_post[1], \"Loss Wait\": sector_choice[\"expected_losses\"][0],\n",
    "        \"Loss Act\": sector_choice[\"expected_losses\"][1], \"Preference\": sector_choice[\"action\"],\n",
    "        \"EVPI\": evpi_from_losses(sector_losses, sector_post),\n",
    "    })\n",
    "sector_summary = pd.DataFrame(sector_rows)\n",
    "print(sector_summary.round(3).to_string(index=False))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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"
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, axes = plt.subplots(2, 2, figsize=(10, 7), sharex=True)\n",
    "for ax, case in zip(axes.flat, sector_cases):\n",
    "    sector, _, _, like_stable, like_adverse, loss_wait, loss_act = case\n",
    "    x = np.linspace(0.02, 0.98, 97)\n",
    "    wait_loss, act_loss = [], []\n",
    "    for p_adverse in x:\n",
    "        beliefs = posterior([1-p_adverse, p_adverse], [like_stable, like_adverse])\n",
    "        losses = expected_losses([[0, loss_wait], [loss_act, 0]], beliefs)\n",
    "        wait_loss.append(losses[0]); act_loss.append(losses[1])\n",
    "    ax.plot(x, wait_loss, label=\"Wait\")\n",
    "    ax.plot(x, act_loss, label=\"Act\")\n",
    "    ax.set_title(sector)\n",
    "    ax.grid(alpha=0.2)\n",
    "axes[1,0].set_xlabel(\"Prior adverse\"); axes[1,1].set_xlabel(\"Prior adverse\")\n",
    "axes[0,0].set_ylabel(\"Expected loss\"); axes[1,0].set_ylabel(\"Expected loss\")\n",
    "axes[0,0].legend()\n",
    "fig.suptitle(\"Sector decision reversals under prior uncertainty\", y=1.02)\n",
    "plt.tight_layout()\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 11 Executive decision record export\n",
    "\n",
    "The export makes assumptions and governance fields reviewable. Analytical preference remains separate from the authorized outcome.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "module                                                                              EP-M03\n",
      "sector                                                                         Agriculture\n",
      "decision                                              Deploy early crop-protection support\n",
      "prior_adverse                                                                          0.4\n",
      "posterior_adverse                                                                     0.64\n",
      "analytical_preference                                                            Intervene\n",
      "expected_loss_margin                                                                  9.92\n",
      "evpi_upper_bound                                                                      2.88\n",
      "evsi_imperfect_test                                                                    0.0\n",
      "sensitivity_note       Preference reverses near prior adverse 0.140 under canonical losses\n",
      "authorized_outcome                                             PENDING HUMAN AUTHORIZATION\n",
      "accountable_owner                                        Programme authorization committee\n",
      "review_trigger              New independent evidence or material change in declared losses\n",
      "\n",
      "CSV preview:\n",
      "module,sector,decision,prior_adverse,posterior_adverse,analytical_preference,expected_loss_margin,evpi_upper_bound,evsi_imperfect_test,sensitivity_note,authorized_outcome,accountable_owner,review_trigger\n",
      "EP-M03,Agriculture,Deploy early crop-protection support,0.4,0.6400000000000001,Intervene,9.920000000000002,2.88,0.0,Preference reverses near prior adverse 0.140 under canonical losses,PENDING HUMAN AUTHORIZATION,Programme authorization committee,New independent evidence or material change in declared losses\n",
      "\n"
     ]
    }
   ],
   "source": [
    "analytical_margin = float(decision[\"expected_losses\"][0] - decision[\"expected_losses\"][1])\n",
    "decision_record = pd.DataFrame([{\n",
    "    \"module\": \"EP-M03\",\n",
    "    \"sector\": \"Agriculture\",\n",
    "    \"decision\": \"Deploy early crop-protection support\",\n",
    "    \"prior_adverse\": prior[1],\n",
    "    \"posterior_adverse\": post[1],\n",
    "    \"analytical_preference\": decision[\"action\"],\n",
    "    \"expected_loss_margin\": analytical_margin,\n",
    "    \"evpi_upper_bound\": evpi,\n",
    "    \"evsi_imperfect_test\": evsi,\n",
    "    \"sensitivity_note\": f\"Preference reverses near prior adverse {reversal_prior:.3f} under canonical losses\",\n",
    "    \"authorized_outcome\": \"PENDING HUMAN AUTHORIZATION\",\n",
    "    \"accountable_owner\": \"Programme authorization committee\",\n",
    "    \"review_trigger\": \"New independent evidence or material change in declared losses\",\n",
    "}])\n",
    "print(decision_record.T.to_string(header=False))\n",
    "csv_buffer = StringIO()\n",
    "decision_record.to_csv(csv_buffer, index=False)\n",
    "print(\"\\nCSV preview:\")\n",
    "print(csv_buffer.getvalue())\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 12 Practice challenges\n",
    "\n",
    "1. Replace the programme prior with the historical base rate. Explain whether the action changes.\n",
    "2. Increase the loss of unnecessary intervention. Find the first value that reverses the action.\n",
    "3. Reduce the reliability of the field signal. Identify the likelihood range over which the action remains stable.\n",
    "4. Add a feasible **Pilot** action with losses between waiting and full intervention.\n",
    "5. For one priority sector, replace the synthetic losses with documented stakeholder assumptions and record their provenance.\n",
    "6. Explain why EVPI is not the amount an institution should automatically spend on more evidence.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "       Stable conditions  Adverse conditions  Posterior expected loss\n",
      "Wait                 0.0                20.0                    12.80\n",
      "Pilot                4.0                 6.0                     5.28\n",
      "Act                  8.0                 0.0                     2.88\n",
      "\n",
      "Preference with a Pilot option: Act\n"
     ]
    }
   ],
   "source": [
    "# Worked challenge: add a bounded Pilot action.\n",
    "pilot_actions = [\"Wait\", \"Pilot\", \"Act\"]\n",
    "pilot_losses = np.array([\n",
    "    [0.0, 20.0],\n",
    "    [4.0, 6.0],\n",
    "    [8.0, 0.0],\n",
    "])\n",
    "pilot_decision = choose_action(pilot_losses, post, pilot_actions)\n",
    "pilot_table = pd.DataFrame(pilot_losses, index=pilot_actions, columns=states)\n",
    "pilot_table[\"Posterior expected loss\"] = pilot_decision[\"expected_losses\"]\n",
    "print(pilot_table.round(3).to_string())\n",
    "print(f\"\\nPreference with a Pilot option: {pilot_decision['action']}\")\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 13 Final verification and human gate\n",
    "\n",
    "The following checks protect the canonical calculations, information-value ordering, four-sector coverage and the separation between analysis and authorization.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "PASS: canonical outputs, uncertainty analyses, sector coverage and human gate verified.\n"
     ]
    }
   ],
   "source": [
    "assert np.allclose(post, [0.36, 0.64])\n",
    "assert np.allclose(decision[\"expected_losses\"], [12.8, 2.88])\n",
    "assert decision[\"action\"] == \"Intervene\"\n",
    "assert np.isclose(evpi, 2.88)\n",
    "assert 0 <= evsi <= evpi + 1e-12\n",
    "assert 0.13 <= reversal_prior <= 0.14\n",
    "assert set(sector_summary[\"Sector\"]) == {\"Agriculture\", \"Education\", \"Health\", \"Habitat\"}\n",
    "assert decision_record.loc[0, \"authorized_outcome\"] == \"PENDING HUMAN AUTHORIZATION\"\n",
    "assert len(likelihood_sensitivity) == 5000\n",
    "print(\"PASS: canonical outputs, uncertainty analyses, sector coverage and human gate verified.\")\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Interpretation and limitations\n",
    "\n",
    "- The examples are synthetic and do not estimate real sector risks.\n",
    "- Priors and likelihoods require provenance, domain review and challenge.\n",
    "- Loss units must not hide rights, legal duties, hard constraints or unequal burdens.\n",
    "- Conditional independence between evidence items is an assumption, not a default truth.\n",
    "- EVPI and EVSI measure potential decision improvement under the declared model; they do not capture every cost, delay or institutional constraint.\n",
    "- Missing states and model misspecification can matter more than numerical precision.\n",
    "- Posterior beliefs inform action but do not authorize it. Human accountability remains decisive.\n"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "name": "python",
   "version": "3"
  },
  "srai": {
   "asset_id": "EP-M03-A08",
   "module": "EP-M03",
   "release": "v0.2",
   "priority_sectors": [
    "Agriculture",
    "Education",
    "Health",
    "Habitat"
   ],
   "public_release_candidate": true,
   "human_authorization_required": true,
   "execution_verified": true,
   "executed_code_cells": 18
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}