2005 | OriginalPaper | Buchkapitel
Modelling the Claim Count with Poisson Regression and Negative Binomial Regression
verfasst von : Bartłomiej Bartoszewicz
Erschienen in: Innovations in Classification, Data Science, and Information Systems
Verlag: Springer Berlin Heidelberg
Aktivieren Sie unsere intelligente Suche, um passende Fachinhalte oder Patente zu finden.
Wählen Sie Textabschnitte aus um mit Künstlicher Intelligenz passenden Patente zu finden. powered by
Markieren Sie Textabschnitte, um KI-gestützt weitere passende Inhalte zu finden. powered by
It is of interest for an insurance company to model the claim count or the claim severity in the presence of covariates or covariate factors describing the policyholders. In this paper the Poisson regression is presented (in the context of generalized linear models) and fitted to car insurance claims data. Since there are symptoms of over-dispersion in the data, the Poisson distribution of the response variable is replaced with the negative binomial distribution and another model is fitted to the number of claims. Finally, a method of testing the significance of the differences between groups of policyholders is shown and applied to the negative binomial regression.