insurance-governance 0.1.1

Unified model governance for UK insurance pricing: PRA SS1/23 validation and model risk management

insurance-governance

Unified model governance for UK insurance pricing teams. Combines PRA SS1/23 statistical validation and model risk management into one package.

Merged from: insurance-validation (PRA SS1/23 validation reports) and insurance-mrm (model risk management).

The problem this solves: validation tests and MRM governance packs were built separately and had separate installs, separate version pinning, and separate import paths. Pricing teams either installed both and managed the coupling themselves, or skipped one. This package resolves that by providing a single install.

Subpackages

insurance_governance.validation

PRA SS1/23 compliant model validation report generator. Runs statistical tests (Gini, PSI, discrimination checks, Hosmer-Lemeshow, lift charts) and produces self-contained HTML reports.

insurance_governance.mrm

Model risk management framework. ModelCard metadata container, RiskTierScorer (objective 0-100 composite score mapping to Tier 1/2/3), ModelInventory (JSON file registry), GovernanceReport (executive committee pack).

Install

pip install insurance-governance

Quick start

import numpy as np
from insurance_governance import (
    ModelValidationReport,
    ValidationModelCard,
    MRMModelCard,
    RiskTierScorer,
    ModelInventory,
    GovernanceReport,
)

# --- Synthetic model outputs (replace with your real model predictions) ---
rng = np.random.default_rng(42)
n_val = 5_000
y_val      = rng.poisson(0.08, n_val).astype(float)   # observed claim counts
y_pred_val = np.clip(rng.normal(0.08, 0.02, n_val), 0.001, None)  # model predictions

# --- Run statistical validation ---
card = ValidationModelCard(
    name="Motor Frequency v3.2",
    version="3.2.0",
    purpose="Predict claim frequency for UK motor portfolio",
    methodology="CatBoost gradient boosting with Poisson objective",
    target="claim_count",
    features=["age", "vehicle_age", "area", "vehicle_group"],
    limitations=["No telematics data"],
    owner="Pricing Team",
)
report = ModelValidationReport(model_card=card, y_val=y_val, y_pred_val=y_pred_val)
report.generate("validation_report.html")

# --- MRM governance pack ---
mrm_card = MRMModelCard(
    model_id="motor-freq-v3",
    model_name="Motor TPPD Frequency",
    version="3.2.0",
    model_class="pricing",
    intended_use="Frequency pricing for private motor.",
)
scorer = RiskTierScorer()
tier = scorer.score(
    gwp_impacted=125_000_000,
    model_complexity="high",
    deployment_status="champion",
    regulatory_use=False,
    external_data=False,
    customer_facing=True,
)
GovernanceReport(card=mrm_card, tier=tier).save_html("mrm_pack.html")

Or import from subpackages directly:

from insurance_governance.validation import ModelValidationReport, ModelCard as ValidationModelCard
from insurance_governance.mrm import ModelCard as MRMModelCard, RiskTierScorer, ModelInventory, GovernanceReport

Note on ModelCard

Both subpackages define a ModelCard class, but they serve different purposes:

• insurance_governance.validation.ModelCard (ValidationModelCard at top level) — Pydantic schema, anchors the statistical validation report, captures features, methodology, limitations.
• insurance_governance.mrm.ModelCard (MRMModelCard at top level) — dataclass, anchors the MRM governance pack, captures assumptions, risk tier, Model Risk Committee metadata.

At the top level they are re-exported as ValidationModelCard and MRMModelCard to avoid ambiguity.

Capabilities Demo

Demonstrated on synthetic motor data: 50,000 UK motor policies, CatBoost Poisson frequency model, 60/20/20 temporal train/validation/test split. Full notebook: notebooks/benchmark.py.

• Runs the full PRA SS1/23 validation suite in a single ModelValidationReport call: Gini coefficient with bootstrap 95% CI, 10-band lift chart, A/E by predicted decile with Poisson CI, Hosmer-Lemeshow goodness-of-fit, PSI on score distribution (train vs validation), monitoring plan completeness check — all returning TestResult objects with a pass/fail flag and human-readable detail
• Computes an overall RAG status (Green/Amber/Red) from the worst-severity failure across all tests
• Produces a self-contained HTML validation report and JSON sidecar, print-to-PDF ready, in under one second
• Scores model risk tier via RiskTierScorer: 6 dimensions (GWP, model complexity, deployment status, regulatory use, external data, customer-facing) mapped to a 0-100 composite with documented rules per point — no subjective judgement required at the MRC presentation
• Registers models in ModelInventory (JSON file, check into git alongside your code); records validation run history linked by run_id; lists overdue reviews
• Generates a GovernanceReport executive committee pack (HTML + JSON) covering model purpose, risk tier rationale, last validation RAG, assumptions register with risk ratings, outstanding issues, approval conditions, and next review date

When to use: You are subject to PRA SS1/23, have 10+ production pricing models, and want consistent, auditable validation and governance output rather than bespoke analyst notebooks that vary by model. Particularly useful before a PRA supervisory visit.

When NOT to use: You need reserving or capital model governance — this package is scoped to pricing models. It also does not replace independent human review of validation results; it automates the tests, not the judgement.

Databricks Notebook

A ready-to-run Databricks notebook benchmarking this library against standard approaches is available in burning-cost-examples.

Performance

Benchmarked on synthetic UK motor data — 50,000 policies, CatBoost Poisson frequency model, 60/20/20 temporal split. See notebooks/benchmark.py for the full demo workflow.

Task	Time
Full SS1/23 validation suite (9 tests)	< 5 seconds
HTML validation report generation	< 1 second
JSON sidecar generation	< 1 second
RiskTierScorer.score()	< 1ms
ModelInventory.register()	< 10ms
GovernanceReport HTML generation	< 1 second

The computational cost of this library is negligible — it runs statistical tests on pre-computed predictions, not raw data. The bottleneck is always the model fitting that produces y_pred_val, not the governance layer. A team generating quarterly validation reports for 15 models can run the full suite in under 10 minutes of wall clock time, most of which is the CatBoost fits.

The value is not in computation speed but in consistency: every model in the portfolio gets identical tests, identical thresholds, and identical output format. A pricing team relying on bespoke analyst notebooks has no guarantee that the Gini in one report means the same thing as the Gini in another.

Related Libraries

Library	What it does
insurance-fairness	Proxy discrimination auditing — fairness audit outputs are a required input to the governance sign-off pack
insurance-deploy	Champion/challenger deployment with ENBP audit logging — governance documents the model; deploy manages its lifecycle
insurance-monitoring	Model monitoring — ongoing monitoring evidence feeds into governance review cycles

Licence

MIT