Non-Life Insurance Pricing with Generalized Linear Models

Non-life insurance pricing is the art of setting the price of an insurance policy, taking into consideration varoius properties of the insured object and the policy holder. Introduced by British actuaries generalized linear models (GLMs) have become today a the standard aproach for tariff analysis.

The book focuses on methods based on GLMs that have been found useful in actuarial practice and provides a set of tools for a tariff analysis. Basic theory of GLMs in a tariff analysis setting is presented with useful extensions of standarde GLM theory that are not in common use.

The book meets the European Core Syllabus for actuarial education and is written for actuarial students as well as practicing actuaries. To support reader real data of some complexity are provided at www.math.su.se/GLMbook.

Written for actuaries and actuariel students Meets the European Core Syllabus for actuarial education Additional real data for deeper understanding available Includes supplementary material: sn.pub/extras

Klappentext
Setting the price of a non-life insurance policy involves the statistical analysis of insurance data, taking into consideration various properties of the insured object and the policy holder. Introduced by British actuaries, generalized linear models (GLMs) have by now become a standard approach used for pricing in many countries. The book focuses on methods based on GLMs that have been found useful in actuarial practice. Basic theory of GLMs in an insurance setting is presented, with useful extensions that are not in common use. The book can be used in actuarial education designed to meet the European Core Syllabus and is written for actuarial students as well as practicing actuaries. To support the readers, it contains case studies using real data of some complexity that are available on the www.

Inhalt
Non-Life Insurance Pricing.- The Basics of Pricing with GLMs.- GLM Model Building.- Multi-Level Factors and Credibility Theory.- Generalized Additive Models.