I. Morillo, L. Bermúdez, D. Karlis

When modelling insurance claim counts data, the actuary often observes overdispersion and an excess of zeros that may be caused by unobserved heterogeneity. A common approach for accounting for overdispersion is to consider models with some overdispersed distribution as opposed to Poisson models. Zero-inflated and compound frequency models are usually applied to insurance data to account for such characteristics of the data. However, a natural way to deal with the unobserved heterogeneity is to consider mixtures of a simpler model. In this paper, we consider K-finite mixtures of some usual regression models. This approach has interesting features: first, it allows for overdispersion and the zero-inflated model represents a special case; and second, it allows for an elegant interpretation based on the typical clustering application of finite mixture models. These models are applied to an automobile insurance claims data set in order to analyse the consequences for risk classification.

Keywords: Zero-inflation, Overdispersion, Finite mixture regression model

Scheduled

PO-1 Poster Session
September 4, 2019  10:40 AM
Multifunctional room. Carbonell building


Other papers in the same session

A bioinformatic study of Single-Cell RNA-seq data analysis protocols for the characterization of cell types of the central nervous system

A. García Galindo, O. González Velasco, J. M. Sánchez Santos, J. De Las Rivas Sanz, E. Sánchez Luis

Análisis de frontera estocástica para estimar eficiencia de centros educativos

C. E. Carleos Artime, N. Corral Blanco, S. Álvarez Morán, A. Shatla

Analizando el índice TGD (Tumor Growth Delay) en la valoración de la eficacia de tratamientos tumorales

P. Román Román, S. Román- Román, J. J. Serrano Pérez, F. Torres Ruiz

Goodness–of–fit tests for quantile regression with missing responses

A. Pérez-González, T. R. Cotos Yáñez, W. González–Manteiga, R. M. Crujeiras

Plugin de R-Commander para la enseñanza de estadística básica: RcmdrPlugin.TeachStat

T. R. Cotos Yáñez, M. A. Mosquera Rodríguez, A. Pérez González, B. Reguengo Lareo


Cookie policy

We use cookies in order to be able to identify and authenticate you on the website. They are necessary for the correct functioning of it, and therefore they can not be disabled. If you continue browsing the website, you are agreeing with their acceptance, as well as our Privacy Policy.

Additionally, we use Google Analytics in order to analyze the website traffic. They also use cookies and you can accept or refuse them with the buttons below.

You can read more details about our Cookie Policy and our Privacy Policy.