← Home

RESEARCH BY INDUSTRY

Evidence for the decision.

Find published research and synthetic demonstrations, with planned and developing topics clearly separated.

9 published public-data studies4 synthetic demonstrations51 planned or developing topics

Insurance

1 published study · 2 planned or developing topics

Published findings & demonstrations

S31 / Insurance

Insurance pricing: interpretable structure versus nonlinear accuracy

An insurance analytics team needs accurate expected-loss estimates that remain understandable across policy segments. Test whether a nonlinear model improves frequency-severity predictions while preserving portfolio and segment calibration.

Evaluated public-data study

Read the finding and evidence ↗
Methods & source

Exposure-offset GLMs; frequency-severity; Tweedie models

freMTPL2 frequency and severity ↗

Research code ↗

Planned topics & work in progress

These topics are separate from completed findings. Each status reflects the work actually performed.

S32 / Insurance

Flood claims and the risk of geographic concentration

An insurer or resilience planner needs to understand tail payments and correlated geographic exposure. Estimate how claim-severity conclusions change when major flood events, coverage limits and portfolio composition are considered.

Planned topic

Proposed design & source

Hierarchical severity; extreme-value analysis; event-level dependence

Clean NFIP claim payments and dated event/geography information. Analyze severity conditional on a reported claim using hierarchical models and carefully diagnosed tail models. For incidence or expected loss per insured exposure, explicitly add compatible redacted policy data and define earned-exposure denominators first.

Evaluation: Hold out entire major events or event-years and geographic groups. Compare empirical severity and lognormal/Gamma baselines with tail models. Report tail quantile calibration, threshold sensitivity and event-bootstrap uncertainty; flag unsupported extrapolation beyond observed experience.

Boundary: Claims alone cannot establish flood probability or claim incidence. Redacted geography, evolving limits and nominal payment amounts complicate comparisons; modeled return periods or portfolio losses need explicit additional assumptions.

OpenFEMA NFIP Redacted Claims ↗

S33 / Insurance

Longevity assumptions and long-horizon liability sensitivity

A benefits or insurance planning team must understand how mortality assumptions affect a long-duration payment obligation. Quantify sensitivity to table choice, longevity improvement and discount rates without pretending the tables are individual training records.

Planned topic

Proposed design & source

Life-contingent valuation; survival probabilities; sensitivity analysis

Select compatible SOA tables with clear population, vintage and select/ultimate or generational definitions. Convert death probabilities into survival curves and expected payment streams. Apply a clearly synthetic cohort of ages and benefits, then compare deterministic and explicitly assumed stochastic improvement scenarios.

Evaluation: Check survival monotonicity, probability identities and actuarial present values against hand-calculated cases. Compare suitable table vintages and run rate/improvement stress grids. Do not report predictive accuracy without separate death and exposure observations.

Boundary: This is a transparent scenario engine using published rates and synthetic obligations. It is not a fitted individual mortality model, an audited reserve calculation or a validated forecast of future longevity.

Mortality and Other Rate Tables (MORT) ↗

The 60-topic program spans 20 industries, with 9 studies published. Synthetic demonstrations are counted separately. Unpublished topics have no public case-study route.

Privacy preferences

Security verification and inquiry delivery are essential. These optional choices do not affect them.

Optional services may remain unavailable until configured. No raw chat, search or contact text belongs in analytics. Read the privacy notice.