Modeling frequency and severity of claims with the zero-inflated generalized cluster-weighted models
From MaRDI portal
Publication:2212142
DOI10.1016/J.INSMATHECO.2020.06.004zbMATH Open1452.91280arXiv1812.11829OpenAlexW3036128637MaRDI QIDQ2212142FDOQ2212142
Authors: Nikola Počuča, Petar Jevtić, Paul D. McNicholas, Tatjana Miljkovic 
Publication date: 19 November 2020
Published in: Insurance Mathematics \& Economics (Search for Journal in Brave)
Abstract: In this paper, we propose two important extensions to cluster-weighted models (CWMs). First, we extend CWMs to have generalized cluster-weighted models (GCWMs) by allowing modeling of non-Gaussian distribution of the continuous covariates, as they frequently occur in insurance practice. Secondly, we introduce a zero-inflated extension of GCWM (ZI-GCWM) for modeling insurance claims data with excess zeros coming from heterogenous sources. Additionally, we give two expectation-optimization (EM) algorithms for parameter estimation given the proposed models. An appropriate simulation study shows that, for various settings and in contrast to the existing mixture-based approaches, both extended models perform well. Finally, a real data set based on French auto-mobile policies is used to illustrate the application of the proposed extensions.
Full work available at URL: https://arxiv.org/abs/1812.11829
Recommendations
- On modeling claim frequency data in general insurance with extra zeros
- Risk classification for claim counts: a comparative analysis of various zero-inflated mixed Poisson and hurdle models
- A new multivariate zero-inflated hurdle model with applications in automobile insurance
- A class of mixture of experts models for general insurance: application to correlated claim frequencies
- Modelling zero-inflated count data with a special case of the generalised Poisson distribution
Classification and discrimination; cluster analysis (statistical aspects) (62H30) Actuarial mathematics (91G05) Applications of statistics to actuarial sciences and financial mathematics (62P05)
Cites Work
- PMTK
- On modeling left-truncated loss data using mixtures of distributions
- Estimating the dimension of a model
- Title not available (Why is that?)
- Title not available (Why is that?)
- Likelihood Ratio Tests for Model Selection and Non-Nested Hypotheses
- Parsimonious mixtures of multivariate contaminated normal distributions
- Title not available (Why is that?)
- The generalized linear mixed cluster-weighted model
- Local statistical modeling via a cluster-weighted approach with elliptical distributions
- Machine learning. A probabilistic perspective
- Zero-Inflated Poisson Regression, with an Application to Defects in Manufacturing
- Clustering and classification via cluster-weighted factor analyzers
- Title not available (Why is that?)
- Cluster-weighted \(t\)-factor analyzers for robust model-based clustering and dimension reduction
- Modeling loss data using mixtures of distributions
- Loss models. From data to decisions
- Fitting mixtures of Erlangs to censored and truncated data using the EM algorithm
- Zero-inflated Poisson regression mixture model
- A mixture likelihood approach for generalized linear models
- Title not available (Why is that?)
- Concomitant variables in finite mixture models
- Model-based clustering via linear cluster-weighted models
- Erratum to: ``The generalized linear mixed cluster-weighted model
- A finite mixture of bivariate Poisson regression models with an application to insurance ratemaking
- Robust clustering in regression analysis via the contaminated Gaussian cluster-weighted model
- Hierarchical insurance claims modeling
- The misuse of the Vuong test for non-nested models to test for zero-inflation
- Lognormal Distributions
- A nested expectation-maximization algorithm for latent class models with covariates
- A class of mixture of experts models for general insurance: theoretical developments
- A class of mixture of experts models for general insurance: application to correlated claim frequencies
- Fat-Tailed Regression Modeling with Spliced Distributions
- A new and intuitive test for zero modification
Cited In (8)
- Model-based clustering via skewed matrix-variate cluster-weighted models
- Risk classification for claim counts: a comparative analysis of various zero-inflated mixed Poisson and hurdle models
- On modeling claim frequency data in general insurance with extra zeros
- Gamma mixture density networks and their application to modelling insurance claim amounts
- Multivariate cluster weighted models using skewed distributions
- Frequency-severity experience rating based on latent Markovian risk profiles
- Prediction of overdispersed count data using real-time cluster-based discretization of explanatory variables
- Finite mixture model of hidden Markov regression with covariate dependence
Uses Software
This page was built for publication: Modeling frequency and severity of claims with the zero-inflated generalized cluster-weighted models
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2212142)
