User menu

Accès à distance ? S'identifier sur le proxy UCLouvain

Fixed versus random effects in Poisson regression models for claim counts: a case study with motor insurance

Primary tabs

Boucher, Jean-Philippe Denuit, Michel [UCL]

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.

Document type Article de périodique (Journal article) – Article de recherche
Publication date 2006
Language Anglais
Journal information "ASTIN Bulletin" - Vol. 36, no. 1, p. 285-301 (2006)
Peer reviewed yes
issn 0515-0361
Publication status Publié
Affiliations Louvain School of Management - CESAM - Center for Studies in Asset Management
UCL - Autre
Keywords fixed effects. ; random effects ; claim frequencies ; risk classification ; Poisson mixture
Links
  1. Allison Paul D., Waterman Richard P., 7. Fixed-Effects Negative Binomial Regression Models, 10.1111/1467-9531.00117
  2. Journal of the Royal Statistical Society B, 32, 283 (1970)
  3. Working Paper (2006)
  4. Regression Analysis of Count Data (1998)
  5. Dean C., Lawless J. F., Willmot G. E., A mixed poisson-inverse-gaussian regression model, 10.2307/3314846
  6. Insurance: Mathematics and Economics, 35, 627 (2004)
  7. Bulletin of the Swiss Association of Actuaries, 51 (2003)
  8. DIONNE Georges, VANASSE Charles, A Generalization of Automobile Insurance Rating Models, 10.2143/ast.19.2.2014909
  9. Dionne G., Vanasse C., Automobile insurance ratemaking in the presence of asymmetrical information, 10.1002/jae.3950070204
  10. Insurance: Mathematics and Economics, 34, 177 (2004)
  11. Hausman Jerry, Hall Bronwyn H., Griliches Zvi, Econometric Models for Count Data with an Application to the Patents-R & D Relationship, 10.2307/1911191
  12. GLIM 82: Proceeding of the International Conference on Generalised Linear Models (1982)
  13. Analysis of Panel Data (2003)
  14. LIANG KUNG-YEE, ZEGER SCOTT L., Longitudinal data analysis using generalized linear models, 10.1093/biomet/73.1.13
  15. Generalized Linear Models (1989)
  16. Mundlak Yair, On the Pooling of Time Series and Cross Section Data, 10.2307/1913646
  17. Biometrika, 68, 563 (1981)
  18. Handbook of Insurance (2000)
  19. Journal of Data Science, 2, 17 (2004)
  20. Tremblay Luc, Using the Poisson Inverse Gaussian in Bonus-Malus Systems , 10.2143/ast.22.1.2005129
  21. Scandinavian Actuarial Journal, 113 (1987)
  22. Econometric Analysis of Count Data (2003)
  23. Insurance: Mathematics and Economics, 36, 153 (2005)
Bibliographic reference Boucher, Jean-Philippe ; Denuit, Michel. Fixed versus random effects in Poisson regression models for claim counts: a case study with motor insurance. In: ASTIN Bulletin, Vol. 36, no. 1, p. 285-301 (2006)
Permanent URL http://hdl.handle.net/2078/16972