Please use this identifier to cite or link to this item:
Fahrmeir, Ludwig
Osuna, Leyre
Year of Publication: 
Series/Report no.: 
Discussion paper // Sonderforschungsbereich 386 der Ludwig-Maximilians-Universität München 334
Overdispersion in count data regression is often caused by neglection or inappropriate modelling of individual heterogeneity, temporal or spatial correlation, and nonlinear covariate effects. In this paper, we develop and study semiparametric count data models which can deal with these issues by incorporating corresponding components in structured additive form into the predictor. The models are fully Bayesian and inference is carried out by computationally efficient MCMC techniques. In a simulation study, we investigate how well the different components can be identified with the data at hand. The approach is applied to a large data set of claim frequencies from car insurance.
Bayesian semiparametric count data regression
negative binomial distribution
Poisson-Gamma distribution
Poisson-Log-Normal distribution
spatial models
Persistent Identifier of the first edition: 
Document Type: 
Working Paper

Files in This Item:
874.54 kB
4.6 MB

Items in EconStor are protected by copyright, with all rights reserved, unless otherwise indicated.