Skip to main content
Log in

Generalized geoadditive models for insurance claims data

Generalisierte geoadditive Modelle zur Schadenanalyse

  • Published:
Blätter der DGVFM

Summary

Generalized regression models provide a flexible framework for analysing insurance claims data. Most applications are still based on generalized linear models, assuming that covariate effects can be modelled by a parametric linear predictor. In many cases, however, the data contain detailed information on metrical and geographical covariates. Their effects are often highly nonlinear, and are at least rather difficult to assess with conventional parametric models. In this paper, we propose generalized geoadditive models which can simultaneously incorporate usual linear effects as well as nonlinear effects of metrical and spatial covariates within a unified semiparametric Bayesian approach. Statistical inference is based on Markov chain Monte Carlo techniques. We apply our methods to analyse the amount of loss and claim frequency for car insurance data from a German insurance company.

Zusammenfassung

Generalisierte Regressionsmodelle ermöglichen die Analyse von Schadenhöhen und Schadenhäufigkeiten im Rahmen eines flexiblen, einheitlichen Ansatzes. Bislang beruhen die meisten Anwendungen auf generalisierten linearen Modellen und damit auf der Annahme, dass die Kovariableneffekte durch einen linearen Prädiktor adäquat erfasst werden können. In vielen Fällen enthalten die Daten jedoch detaillierte Information über metrische und räumlich-geographische Kovariablen, deren Effekte oft deutlich nichtlinear und mit konventionellen parametrischen Modellen meist schwierig zu analysieren sind. Wir entwickeln generalisierte geoadditive Modelle, für die übliche lineare Effekte sowie nichtlineare Effekte von metrischen und räumlichen Kovariablen simultan mit einem semiparametrischen Bayes-Ansatz analysiert werden können. Die statistische Inferenz basiert auf Markov-Ketten Monte Carlo (MCMC) Techniken. Wir wenden die Methodik zur Analyse von Schadenhöhen und -häufigkeiten von KFZ-Versicherungsdaten eines deutschen Versicherungsnehmers an.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Similar content being viewed by others

References

  • Besag, J.;J. York;A. Mollie (1991): Bayesian image restoration with two applications in spatial statistics (with discussion). Annals of the Institute of Statistical Mathematics, 43, 1–59.

    Article  MathSciNet  MATH  Google Scholar 

  • Besag, J.;C. Kooperberg, (1995): On Conditional and Intrinsic Autoregressions. Biometrika, 82, 733–746.

    MathSciNet  MATH  Google Scholar 

  • Brezger, A.; S. Lang, S. (2003): Generalized additive regression based on Bayesian P-splines. SFB 386 Discussion paper 321, Department of Statistics, University of Munich.

  • Brockmann, M.;S. Wright (1992): Statistical Motor Rating: Making Effective Use of Your Data. Journal of the Institute of Actuaries, 119, 457–543.

    Article  Google Scholar 

  • Eilers, P. H. C.;B. D. Marx (1996): Flexible smoothing using B-splines and penalized likelihood (with comments and rejoinder). Statistical Science, 11 (2), 89–121.

    Article  MathSciNet  MATH  Google Scholar 

  • Fahrmeir, L.;S. Lang (2001): Bayesian Inference for Generalized Additive Mixed Models Based on Markov Random Field Priors. Journal of the Royal Statistical Society C (Appl. Stat.), 50, 201–220.

    Article  MathSciNet  Google Scholar 

  • Fahrmeir, L.;S. Lang (2001): Bayesian Semiparametric Regression Analysis of Multicategorical Time-Space Data. Annals of the Institute of Statistical Mathematics, 53, 10–30.

    Article  MathSciNet  MATH  Google Scholar 

  • Fahrmeir, L.;G. Tutz (2001): Multivariate Statistical Modelling based on Generalized Linear Models, Springer-Verlag, New York.

    Book  MATH  Google Scholar 

  • Haberman, S.;A. Renshaw (1998): Actuarial Applications of Generalized Linear Models. In D. Hand and S. Jacka (Eds.), Statistics in Finance. Arnold, London.

    Google Scholar 

  • Hastie T.;R. Tibshirani (1990): Generalized Additive Models. Chapman and Hall, London.

    MATH  Google Scholar 

  • Jørgensen, B. (1997): The Theory of Dispersion Models. Chapman and Hall, London.

    MATH  Google Scholar 

  • Lang, S.; A. Brezger (2003): Bayesian P-splines. Journal of Computational and Graphical Statistics, to appear.

  • Lang, S.; P. Kragler; G. Haybach; L. Fahrmeir (2002): Bayesian Space-Time Analysis of Health Insurance Data. In: Exploratory Data Analysis in Empirical Research. Edited by M. Schwaiger, and O. Opitz

  • Mack, T. (1991): A Simple Parametric Model for Rating Automobile Insurance or Estimating IBNR Claims Reserves. Astin Bulletin, 22, 93.

    Google Scholar 

  • Mack, T. (1998): Schadensversicherungsmathematik. Vol. 28 of Schriftenreihe Angewandte Versicherungsmathematik. Verlag Versicherungswirtschaft, Karlsruhe.

    Google Scholar 

  • Renshaw, A. (1994): Modelling the Claims Process in the Presence of Covariates. Astin Bulletin, 24, 265–285.

    Article  Google Scholar 

  • Winkelmann, R. (1997): Econometric Analysis of Count Data (2nd ed.). Springer, Berlin.

    Book  MATH  Google Scholar 

Download references

Author information

Authors and Affiliations

Authors

Rights and permissions

Reprints and permissions

About this article

Cite this article

Fahrmeir, L., Lang, S. & Spies, F. Generalized geoadditive models for insurance claims data. Blätter DGVFM 26, 7–23 (2003). https://doi.org/10.1007/BF02808770

Download citation

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/BF02808770

Keywords

Navigation