Implementation of Frequency-Severity Association in BMS Ratemaking

A Bonus-Malus System (BMS) in insurance is a premium adjustment mechanism widely used in a posteriori ratemaking process to set the premium for the next contract period based on a policyholder's claim history. The current practice in BMS implementation relies on the assumption of independence between claim frequency and severity, despite the fact that a series of recent studies report evidence of a significant frequency-severity relationship, particularly in automobile insurance. To address this discrepancy, we propose a copula-based correlated random effects model to accommodate the dependence between claim frequency and severity, and further illustrate how to incorporate such dependence into the current BMS. We derive analytical solutions to the optimal relativities under the proposed framework and provide numerical experiments based on real data analysis to assess the effect of frequency-severity dependence in BMS ratemaking.

[2]  Christian Genest,et al.  Generalized linear models for dependent frequency and severity of insurance claims , 2015 .

[3]  C. Borror.  Generalized Linear Models and Extensions, Second Edition , 2008 .

[4]  Charles E McCulloch,et al.  Estimation of covariate effects in generalized linear mixed models with informative cluster sizes. , 2011, Biometrika.

[5]  E. Gómez-Déniz,et al.  The net Bayes premium with dependence between the risk profiles , 2009 .

[6]  Uditha Balasooriya,et al.  Optimal relativities and transition rules of a bonus–malus system , 2015 .

[7]  Gillian Z. Heller,et al.  Generalized Linear Models for Insurance Data , 2008 .

[8]  Peng Shi,et al.  Dependent frequency–severity modeling of insurance claims , 2015 .

[9]  Jean Pinquet,et al.  Allowance for Cost of Claims in Bonus-Malus Systems , 1997, ASTIN Bulletin.

[10]  C. Czado,et al.  A mixed copula model for insurance claims and claim sizes , 2012 .

[11]  Bradley P. Carlin,et al.  Bayesian measures of model complexity and fit , 2002 .

[12]  Lutz F. Gruber,et al.  Bayesian total loss estimation using shared random effects , 2015 .

[13]  Jae Youn Ahn,et al.  Does hunger for bonuses drive the dependence between claim frequency and severity? , 2018, Insurance: Mathematics and Economics.

[14]  David Landriault,et al.  On a risk model with dependence between interclaim arrivals and claim sizes , 2006 .

[15]  Martyn Plummer,et al.  JAGS: A program for analysis of Bayesian graphical models using Gibbs sampling , 2003 .

[16]  Gee Y. Lee,et al.  Multivariate Frequency-Severity Regression Models in Insurance , 2016, Risks.

[17]  Ruodu Wang,et al.  Collective Risk Models with Dependence Uncertainty , 2017 .

[18]  Michel Denuit,et al.  Setting a Bonus-Malus Scale in the Presence of Other Rating Factors , 2003 .