Construction of Life Boards of invalids using Bayesian statistics

Abstract

The objective of this article is to construct life tables of invalids among urban beneficiaries of the General Regime of the National Brazilian Social Security System. Assuming that the number of deaths in each table follows a Poisson distribution, the death rates by simple age were graduated through Bayesian statistical methods, by using the Gompertz-Makehan parametric model and using Bayesian statistical inferences. Life tables were constructed for death rates of men and women, and credibility intervals were set up for the parameters and components of the model, as well as for the death rates and functions of the board. An application was made by calculating an annuity and the actuarial liabilities.