Abstract:
This paper presents a new method of the parametric graduation of mortality rate which uses Bayesian Reversible Jump Markov Chain Monte Carlo methods. The new method can be seen as an automatic graduation method which this graduation method deals satisfactorily with the data in each case, without the need for any intervention from the graduator. For comparison, we also apply graduation using generalized linear models to the same mortality rates. Mortality data using in this study are life insured mortality rates, Thai populations mortality rates and mortality rates from Monte Carlo simulation. The suitable graduation methods were chosen by considering the smallest value of the mean absolute percent error (MAPE). The results of the study show that the MAPE of graduation using generalized linear models are smaller than those of graduation using Bayesian trans-dimensional models. However, automated graduation using Bayesian trans-dimensional models can be adjusted the mortality rate to the law of mortality.