Heterogeneous INAR(1) model with application to car insurance

Abstract The bonus-malus scheme shows how the history of claim arrivals determines the dynamics of insurance premium. It is important to distinguish to what extent changes in the insurance premium are explained by lagged claim counts introduced among explanatory variables and by unobservable heterogeneity included in the model, which needs to be regularly updated. For this purpose, we introduce the integer valued autoregressive (INAR) model with unobserved heterogeneity. The model is applied to premium updating in car insurance and compared to the standard method based on the negative binomial distribution. We find that the premium depends on the claim history and that the timing of claim arrivals matters. This result is different from the outcome of the standard framework in which the average number of claims per year is the only relevant factor.

[1]  Eric R. Ziegel,et al.  Generalized Linear Models , 2002, Technometrics.

[2]  Georges Dionne,et al.  Handbook of insurance , 2000 .

[3]  Mohamed Alosh,et al.  FIRST‐ORDER INTEGER‐VALUED AUTOREGRESSIVE (INAR(1)) PROCESS , 1987 .

[4]  U. Böckenholt.  Mixed INAR(1) Poisson regression models: Analyzing heterogeneity and serial dependencies in longitudinal count data , 1998 .

[5]  C. Gouriéroux,et al.  12 Pseudo-likelihood methods , 1993 .

[6]  P. Diggle,et al.  Analysis of Longitudinal Data. , 1997 .

[7]  A. Cameron,et al.  Econometric models based on count data. Comparisons and applications of some estimators and tests , 1986 .

[8]  Christian Gourieroux,et al.  Nonlinear Panel Data Models with Dynamic Heterogeneity , 1998 .

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

[10]  Jean Lemaire,et al.  Automobile Insurance: Actuarial Models , 1985 .

[11]  Arthur Snow,et al.  Evidence on Adverse Selection: Equilibrium Signaling and Cross-Subsidization in the Insurance Market , 1994, Journal of Political Economy.

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

[13]  Ed. McKenzie,et al.  SOME SIMPLE MODELS FOR DISCRETE VARIATE TIME SERIES , 1985 .

[14]  J. Lawless.  Negative binomial and mixed Poisson regression , 1987 .

[15]  Montserrat Guillén,et al.  Allowance for the Age of Claims in Bonus-Malus Systems* , 2001, ASTIN Bulletin.

[16]  E. McKenzie,et al.  Some ARMA models for dependent sequences of poisson counts , 1988, Advances in Applied Probability.

[17]  P. McCullagh.  Quasi-Likelihood Functions , 1983 .

[18]  Bjørn Sundt.  Credibility estimators with geometric weights , 1988 .

[19]  Christian Gourieroux,et al.  A count data model with unobserved heterogeneity , 1997 .

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

[21]  W. Neuhaus.  A bonus—malus system in automobile insurance , 1988 .

[22]  C. Gouriéroux.  The Econometrics of Risk Classification in Insurance , 1999 .

[23]  Mohamed Alosh,et al.  An integer-valued pth-order autoregressive structure (INAR(p)) process , 1990, Journal of Applied Probability.

[24]  Kurt Brännäs,et al.  GENERALIZED INTEGER-VALUED AUTOREGRESSION , 2001 .

[25]  M. Arellano.  Panel Data Econometrics , 2002 .

[26]  Christian Gourieroux,et al.  Testing for Evidence of Adverse Selection in the Automobile Insurance Market: A Comment , 2001, Journal of Political Economy.

[27]  Pierre-André Chiappori,et al.  Testing for Asymmetric Information in Insurance Markets , 2000, Journal of Political Economy.

[28]  C. Gouriéroux,et al.  Evidence of Adverse Selection in Automobile Insurance Markets , 1998 .

[29]  C. Gourieroux,et al.  Pseudo Maximum Likelihood Methods: Applications to Poisson Models , 1984 .

[30]  Christian Gourieroux,et al.  Simulation-based econometric methods , 1996 .

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

[32]  C. Gouriéroux,et al.  PSEUDO MAXIMUM LIKELIHOOD METHODS: THEORY , 1984 .