Actuarial Python Packages

chainladder

Actuarial reserving in Python

rustystats

Fast Generalized Linear Models with a Rust backend

modelx

Use Python like a spreadsheet!

actuarialpy

Core actuarial analysis tools for Python: experience analysis, per-exposure metrics, trend, completion, seasonality, credibility, exposure, and TVM.

lossmodels

Actuarial loss modeling in Python: severity, frequency, aggregate loss, fitted distributions, and insurance loss analytics.

ratingmodels

Pricing and rate-indication tools for Python: manual rating, experience rating, GLM relativities, retention gross-up, renewal, and audit trails.

aggregate

Tools for creating and working with aggregate probability distributions.

extremeloss

Extreme-value tail-risk estimation in Python: peaks-over-threshold, GPD fitting, rare-event analysis, and large-claim loading.

risksim

Portfolio Monte Carlo risk simulation in Python for insurance risk, aggregate loss, capital modeling, and risk measures.

jact

JAX-based transition probability computation for multi-state models

actuarialmath

Python package to solve actuarial life-contingent risks

haute

Open-source pricing engine

elizur

Elizur is a finance library for actuaries, finance professionals, and students

insurance-monitoring

Model drift detection, calibration monitoring, and sequential A/B testing for insurance pricing models: PSI, Gini drift, Murphy decomposition, mSPRT champion/challenger.

insurance-causal

Causal inference for insurance pricing: double machine learning, price elasticity, heterogeneous treatment effects, and post-hoc rate change evaluation for UK personal lines.

insurance-datasets

Synthetic UK insurance datasets with known data generating processes

insurance-fairness

Proxy discrimination auditing, discrimination-free pricing, and fairness diagnostics for UK insurance pricing models under FCA Consumer Duty and Equality Act 2010.

morai

Actuarial experience dashboard and predictor

insurance-causal-policy

Causal evaluation of insurance pricing interventions using Synthetic Difference-in-Differences

insurance-gam

Interpretable GAM toolkit for insurance pricing: EBM, Neural Additive Models, Pairwise Interaction Networks, and debiased post-selection GLM inference for UK actuarial modelling.

insurance-conformal

Distribution-free prediction intervals for insurance pricing: conformal coverage guarantees, Tweedie non-conformity scores, SCR bounds, and anytime-valid sequential monitoring.

insurance-quantile

Actuarial tail risk quantile and expectile regression for UK personal lines pricing, including EQRN extreme quantile neural nets (Pasche & Engelke 2024)

shap-relativities

SHAP-based rating relativities from GBM models for insurance pricing — extract GLM-style factors from CatBoost, Polars

lossratio

Loss ratio analysis and projection for long-term health insurance. In development.