Fixed versus Random Effects in Poisson Regression Models for Claim Counts: A Case Study with Motor Insurance

This paper examines the validity of some stylized statements that can be found in the actuarial literature about random effects models. Specifically, the actual meaning of the estimated parameters and the nature of the residual heterogeneity are discussed. A numerical illustration performed on a Belgian motor third party liability portfolio supports this discussion.

[1]  S. Zeger,et al.  Longitudinal data analysis using generalized linear models , 1986 .

[2]  Y. Mundlak.  On the Pooling of Time Series and Cross Section Data , 1978 .

[3]  P. McCullagh,et al.  Generalized Linear Models , 1992 .

[4]  J. T. Wulu,et al.  Regression analysis of count data , 2002 .

[5]  P. McCullagh,et al.  Generalized Linear Models , 1972, Predictive Analytics.

[6]  Michel Denuit,et al.  Non-life rate-making with Bayesian GAMs , 2004 .

[7]  Cheng Hsiao,et al.  Analysis of Panel Data , 1987 .

[8]  Michel Denuit,et al.  Tarification automobile sur données de panel , 2003 .

[9]  R. Winkelmann.  Econometric Analysis of Count Data , 1997 .

[10]  Luc Tremblay.  Using the Poisson Inverse Gaussian in Bonus-Malus Systems , 1992, ASTIN Bulletin.

[11]  Juni Palmgren,et al.  The Fisher information matrix for log linear models arguing conditionally on observed explanatory variable , 1981 .

[12]  E. B. Andersen,et al.  Asymptotic Properties of Conditional Maximum‐Likelihood Estimators , 1970 .

[13]  John Hinde,et al.  Compound Poisson Regression Models , 1982 .

[14]  Michel Denuit,et al.  Risk Classification for Claim Counts , 2007 .

[15]  Gordon E. Willmot,et al.  A mixed poisson–inverse‐gaussian regression model , 1989 .

[16]  G. Dionne,et al.  Automobile Insurance Ratemaking In The Presence Of Asymmetric Information , 1992 .

[17]  G. Dionne,et al.  A Generalization of Automobile Insurance Rating Models: The Negative Binomial Distribution with a Regression Component , 1988, ASTIN Bulletin.

[18]  Gordon E. Willmot,et al.  The Poisson-Inverse Gaussian distribution as an alternative to the negative binomial , 1987 .

[19]  Pravin K. Trivedi,et al.  Regression Analysis of Count Data , 1998 .

[20]  P. Allison,et al.  7. Fixed-Effects Negative Binomial Regression Models , 2002 .

[21]  Kelvin K. W. Yau,et al.  On modeling claim frequency data in general insurance with extra zeros , 2005 .

[22]  Z. Griliches,et al.  Econometric Models for Count Data with an Application to the Patents-R&D Relationship , 1984 .

[23]  Jean Pinquet,et al.  Experience Rating through Heterogeneous Models , 2000 .

[24]  Z. Griliches,et al.  Econometric Models for Count Data with an Application to the Patents-R&D Relationship , 1984 .

[25]  Musa H. Asyali,et al.  The Poisson Inverse Gaussian Regression Model in the Analysis of Clustered Counts Data , 2021, Journal of Data Science.

[26]  Christian Gourieroux,et al.  Heterogeneous INAR(1) model with application to car insurance , 2004 .