A CLASS OF MIXTURE OF EXPERTS MODELS FOR GENERAL INSURANCE: APPLICATION TO CORRELATED CLAIM FREQUENCIES

This paper focuses on the estimation and application aspects of the Erlang count logit-weighted reduced mixture of experts model (EC-LRMoE), which is a fully flexible multivariate insurance claim frequency regression model. We first prove the identifiability property of the proposed model to ensure...

Full description

Saved in:
Bibliographic Details
Published in:ASTIN Bulletin : The Journal of the IAA Vol. 49; no. 3; pp. 647 - 688
Main Authors: Chai Fung, Tsz, Badescu, Andrei L., Sheldon Lin, X.
Format: Journal Article
Language:English
Published: New York, USA Cambridge University Press 01.09.2019
Subjects:
Actuarial science
Algorithms
Calibration
Datasets
Insurance claims
Insurance premiums
Parameter estimation
Policyholders
Statistical inference
expectation conditional maximization algorithm
multivariate count regression
logit-weighted gating functions
Erlang count models
mixture of experts models
ISSN:0515-0361, 1783-1350
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:This paper focuses on the estimation and application aspects of the Erlang count logit-weighted reduced mixture of experts model (EC-LRMoE), which is a fully flexible multivariate insurance claim frequency regression model. We first prove the identifiability property of the proposed model to ensure that it is a suitable candidate for statistical inference. An expectation conditional maximization (ECM) algorithm is developed for efficient model calibrations. Three simulation studies are performed to examine the effectiveness of the proposed ECM algorithm and the versatility of the proposed model. The applicability of the EC-LRMoE is shown through fitting an European automobile insurance data set. Since the data set contains several complex features, we find it necessary to adopt such a flexible model. Apart from showing excellent fitting results, we are able to interpret the fitted model in an insurance perspective and to visualize the relationship between policyholders’ information and their risk level. Finally, we demonstrate how the fitted model may be useful for insurance ratemaking.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:0515-0361
1783-1350
DOI:10.1017/asb.2019.25