Ioannis Ntzoufras

Publications Page

Bayesian Modelling of Outstanding Liabilities
Incorporating Claim Count Uncertainty 

I. Ntzoufras and Petros Dellaportas (2002)

North American Actuarial Journal, 6, 113-128. 

Abstract

        This paper deals with the prediction of the amount of outstanding automobile claims that an insurance company will pay in the near future. We consider various competing models using Bayesian theory and Markov chain Monte Carlo methods. Claim counts are used in order to add a further hierarchical stage in the model with log-normally distributed claim amounts and its corresponding state space version. By this way, we incorporate information from both the outstanding claim amounts and counts data resulting to new model formulations. We provide implementation details and illustrations with real insurance data.


Keywords:  Log-normal model; Markov chain Monte Carlo; Missing values; Multinomial model; State space models.

[Download Paper]

[ Download the correct Table 3.]

[Download Ph.D. case study with detailed computations]


Back to Research Page
All Contents Copyright.
Last revised: 15-11-2001