Approximate Bayesian computations to fit and compare insurance loss models

DOI10.1016/J.INSMATHECO.2021.06.002zbMATH Open1471.91459arXiv2007.03833OpenAlexW3182517088MaRDI QIDQ2234770FDOQ2234770

Authors: Patrick J. Laub, Pierre-Olivier Goffard 

Publication date: 19 October 2021

Published in: Insurance Mathematics \& Economics (Search for Journal in Brave)

Abstract: Approximate Bayesian Computation (ABC) is a statistical learning technique to calibrate and select models by comparing observed data to simulated data. This technique bypasses the use of the likelihood and requires only the ability to generate synthetic data from the models of interest. We apply ABC to fit and compare insurance loss models using aggregated data. A state-of-the-art ABC implementation in Python is proposed. It uses sequential Monte Carlo to sample from the posterior distribution and the Wasserstein distance to compare the observed and synthetic data.

Full work available at URL: https://arxiv.org/abs/2007.03833

Recommendations

• New insights into approximate Bayesian computation
• Modelling extremes using approximate Bayesian computation
• Approximate Bayesian computational methods
• An Introduction to Approximate Bayesian Computation
• Approximate Bayesian computation with differential evolution

zbMATH Keywords

sequential Monte Carloapproximate Bayesian computationBayesian statisticslikelihood-free inferenceWasserstein distancecompound distributionstatistical claim modeling

Mathematics Subject Classification ID

Actuarial mathematics (91G05) Applications of statistics to actuarial sciences and financial mathematics (62P05) Numerical methods (including Monte Carlo methods) (91G60)

Cites Work

• Likelihood-based and Bayesian methods for Tweedie compound Poisson linear mixed models
• Bayesian data analysis.
• Multilevel modeling of insurance claims using copulas
• Sequential Imputations and Bayesian Missing Data Problems
• Bayes Factors
• Approximate Bayesian computation: a nonparametric perspective
• Loss models. From data to decisions
• A simple approach to maximum intractable likelihood estimation
• Ruin probabilities
• Adaptive approximate Bayesian computation
• Handbook of approximate Bayesian computation
• ABC likelihood-free methods for model choice in Gibbs random fields
• An adaptive sequential Monte Carlo method for approximate Bayesian computation
• Semi-automatic selection of summary statistics for ABC model choice
• Fitting Tweedie's compound poisson model to insurance claims data
• Fitting Tweedie's Compound Poisson Model to Insurance Claims Data: Dispersion Modelling
• On parameter estimation with the Wasserstein distance
• Decompounding: an estimation problem for Poisson random sums.
• A kernel type nonparametric density estimator for decompounding
• Claims Reserving Using Tweedie's Compound Poisson Model
• Dependent frequency-severity modeling of insurance claims
• Chain ladder method: Bayesian bootstrap versus classical bootstrap
• A non-parametric Bayesian approach to decompounding from high frequency data
• Efficient nonparametric inference for discretely observed compound Poisson processes
• Generalised linear models for aggregate claims: to Tweedie or not?
• Generalized linear models for dependent frequency and severity of insurance claims
• Efficient and accurate approximate Bayesian inference with an application to insurance data
• On probability distributions of present values in life insurance
• Approximate Bayesian Computation with the Wasserstein Distance

Cited In (6)

• Discrepancy-based inference for intractable generative models using quasi-Monte Carlo
• New challenges in the interplay between finance and insurance. Abstracts from the workshop held October 1--6, 2023
• Sequential Monte Carlo samplers to fit and compare insurance loss models
• Quantification of automobile insurance liability: A Bayesian failure time approach.
• Cyber risk frequency, severity and insurance viability
• Method of Winsorized Moments for Robust Fitting of Truncated and Censored Lognormal Distributions

Uses Software

• CASdatasets
• PyMC
• BayesDA
• Python

This page was built for publication: Approximate Bayesian computations to fit and compare insurance loss models

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2234770)