{
  "runId": "experience-engine-1.1:fde472467624a1589b09a7392b98264b17610d831f587d97fa95f5b0da22789a",
  "revision": "fde472467624a1589b09a7392b98264b17610d831f587d97fa95f5b0da22789a",
  "source": {
    "analysisId": "cmuo7toaj0002jy04nryugb5h",
    "datasetKey": "public-evidence-only",
    "datasetName": "Public evidence only / no CaseSet",
    "provenance": "UNKNOWN / DILIGENCE REQUIRED"
  },
  "engineVersion": "experience-engine-1.1",
  "inputSnapshot": [],
  "report": {
    "engineVersion": "experience-engine-1.1",
    "profile": {
      "total": 0,
      "eligible": 0,
      "excluded": 0,
      "duplicateIdentityRows": 0,
      "groups": 0,
      "decisionTypes": 0,
      "linked": 0,
      "economicValueRows": 0,
      "missing": {
        "group": 0,
        "caseType": 0,
        "decision": 0,
        "action": 0,
        "outcome": 0,
        "grade": 0,
        "decisionTime": 0,
        "outcomeTime": 0
      },
      "invalidChronology": 0
    },
    "protocol": {
      "task": "Predict CORRECT versus other resolved grades",
      "features": [
        "decisionClassId/sourceRecordType",
        "caseType",
        "customerSegment (local/blended only)"
      ],
      "training": 0,
      "validation": 0,
      "finalTest": 0,
      "split": "Chronological 60/20/20; tied timestamps stay together; feedback must precede prediction; final-test labels never select a model.",
      "selection": "Lowest validation Brier; ties prefer the simpler model. Fixed Beta(1,1) base prior and four-case pattern shrinkage.",
      "inference": "Descriptive estimator comparisons; no significance or causal claim. Final-test comparisons and learning curves are reported together, not cherry-picked."
    },
    "ready": false,
    "blockedReason": "Need at least 20 training, 10 validation and 20 final-test cases with chronologically available grades; found 0/0/0. At least 10 grades must be available before validation and 20 before final evaluation.",
    "validationScores": [],
    "selected": null,
    "scores": [],
    "curve": [],
    "segmentResults": [],
    "economicByGrade": [
      {
        "grade": "CORRECT",
        "count": 0,
        "mean": null
      },
      {
        "grade": "PARTIALLY_CORRECT",
        "count": 0,
        "mean": null
      },
      {
        "grade": "INCORRECT",
        "count": 0,
        "mean": null
      }
    ],
    "findings": [
      {
        "id": "capture",
        "family": "EXPERIENCE_CAPTURE",
        "status": "BLOCKED",
        "summary": "0/0 cases contain linked decisions, actions, outcomes and resolved grades.",
        "interpretation": "Measures usable feedback coverage; grade reliability requires independent regrading.",
        "nextTest": "Audit incomplete rows and independently regrade a sample.",
        "sample": 0,
        "metrics": {
          "complete": 0,
          "missing": 0,
          "duplicateIdentityRows": 0
        }
      },
      {
        "id": "predictability",
        "family": "LEARNING_CAUSALITY",
        "status": "BLOCKED",
        "summary": "Need at least 20 training, 10 validation and 20 final-test cases with chronologically available grades; found 0/0/0. At least 10 grades must be available before validation and 20 before final evaluation.",
        "interpretation": "Tests whether past experience predicts grades on later cases. It does not measure improvement from choosing different actions.",
        "nextTest": "Compare decisions made with and without the selected learner on a new evaluation batch.",
        "sample": 0,
        "metrics": {
          "selected": null,
          "selectedBrier": null,
          "baselineBrier": null,
          "gain": null,
          "finalSetLeader": null,
          "selectionReversal": "NO"
        }
      },
      {
        "id": "learning-curve",
        "family": "MARGINAL_INFORMATION_VALUE",
        "status": "BLOCKED",
        "summary": "Need at least 20 training, 10 validation and 20 final-test cases with chronologically available grades; found 0/0/0. At least 10 grades must be available before validation and 20 before final evaluation.",
        "interpretation": "A fixed-test learning curve measures whether additional historical cases help this estimator; results may be non-monotonic.",
        "nextTest": "Repeat on a fresh time window and examine changing case mix.",
        "sample": 0,
        "metrics": {
          "firstBrier": null,
          "lastBrier": null,
          "directionReversal": "NO"
        }
      },
      {
        "id": "reconstruction-proxy",
        "family": "REBUILDABILITY_COMPRESSION",
        "status": "BLOCKED",
        "summary": "Reconstruction has not been tested: no challenger with restricted data access or a calibration budget has been evaluated.",
        "interpretation": "The grade-prediction models share historical data access. Their scores cannot establish how cheaply a competitor could reproduce the capability.",
        "nextTest": "Compare against a challenger restricted to public features, rules and a fixed calibration budget.",
        "sample": 0,
        "metrics": {
          "challengerCases": 0
        }
      },
      {
        "id": "economics",
        "family": "ECONOMIC_MATERIALITY",
        "status": "BLOCKED",
        "summary": "0/0 cases carry finite economic values; mean recorded value unavailable.",
        "interpretation": "Grade-stratified values describe recorded stakes. Units and currency must be consistent before aggregation; these are not causal savings estimates.",
        "nextTest": "Specify the value unit and compare policy outcomes under matched evaluation conditions.",
        "sample": 0,
        "metrics": {
          "CORRECT": null,
          "PARTIALLY_CORRECT": null,
          "INCORRECT": null
        }
      },
      {
        "id": "time-shift",
        "family": "LEARNING_CAUSALITY",
        "status": "BLOCKED",
        "summary": "Correct-grade rate: early unavailable, late unavailable.",
        "interpretation": "Descriptive time shift may reflect case mix, policy or grading changes; it is not attributed to learning.",
        "nextTest": "Compare matched decision types across time and identify policy/model versions.",
        "sample": 0,
        "metrics": {
          "firstRate": null,
          "lastRate": null
        }
      }
    ],
    "transfer": {
      "eligible": 0,
      "excluded": 0,
      "segments": 0,
      "patternCount": 0,
      "sharedPatterns": 0,
      "sharedCases": 0,
      "training": 0,
      "heldOut": 0,
      "scored": 0,
      "local": null,
      "pooled": null,
      "delta": null,
      "bySegment": [],
      "enough": false,
      "finding": "Insufficient held-out coverage",
      "method": "Segment transfer probe v1: earliest 70% train, latest 30% test; only feedback available before each test decision is used. Predict CORRECT versus other resolved grades from decision class and case type. Same fixed smoothing in each arm; lower Brier score is better. A 0.005 Brier difference defines the descriptive effect band, not statistical significance.",
      "scope": "Groups are customer segments, not identified customers. This tests transfer of grade prediction; improving chosen actions requires an action-policy experiment."
    },
    "predictions": [],
    "nextExperiments": [
      {
        "name": "Action-policy trial",
        "required": "Identified available actions plus randomized assignment, logged propensities, or an explicit reward simulator",
        "status": "REQUIRES_DESIGN"
      },
      {
        "name": "Customer-level transfer",
        "required": "Verified customer IDs, permitted pooling scope and a held-out customer protocol",
        "status": "REQUIRES_DESIGN"
      },
      {
        "name": "Budgeted challenger",
        "required": "Competitor-access feature set and calibration budget",
        "status": "REQUIRES_DESIGN"
      }
    ]
  }
}