Statistical models for an MTPL portfolio

Daria Pirozhkova

Master's thesis

Abstract:

In this thesis, we consider several statistical techniques applicable to claim frequency models of an MTPL portfolio with a focus on overdispersion. The practical part of the work is focused on the application and comparison of the models on real data represented by an MTPL portfolio. The comparison is presented by the results of goodness-of-fit measures. Furthermore, the predictive power of selected models is tested for the given dataset, using the simulation method. Hence, this thesis provides a combination of the analysis of goodness-of-fit results and the predictive power of the models.

Abstract:

In this thesis, we consider several statistical techniques applicable to claim frequency models of an MTPL portfolio with a focus on overdispersion. The practical part of the work is focused on the application and comparison of the models on real data represented by an MTPL portfolio. The comparison is presented by the results of goodness-of-fit measures. Furthermore, the predictive power of selected models is tested for the given dataset, using the simulation method. Hence, this thesis provides a combination of the analysis of goodness-of-fit results and the predictive power of the models.

Keywords

pricing model, claim frequency, Negative Binomial model, Overdispersed Poisson model, Poisson model, count data model, overdispersion, cross-validation, predictive power, goodness-of-fit

Keywords

Poisson model, overdispersion, cross-validation, predictive power, Overdispersed Poisson model, claim frequency, pricing model, goodness-of-fit, Negative Binomial model, count data model

Language used: English

Date on which the thesis was submitted / produced: 26. 10. 2016

Thesis defence

• Date of defence: 8. 6. 2017
• Supervisor: Pavel Zimmermann
• Reader: Ivana Malá

Vysoká škola ekonomická v Praze

Master programme / field:
Kvantitativní metody v ekonomice / Quantitative Economic Analysis