Modeling future mortality rates using both deterministic and stochastic approaches


Tezin Türü: Yüksek Lisans

Tezin Yürütüldüğü Kurum: Orta Doğu Teknik Üniversitesi, Uygulamalı Matematik Enstitüsü, Aktüerya Bilimleri Anabilim Dalı, Türkiye

Tezin Onay Tarihi: 2015

Öğrenci: ETKİN HASGÜL

Danışman: SEVTAP AYŞE KESTEL

Özet:

This study is based on creating an alternative mortality rates projection method for constructingthefuturemortalitytablesofseveralcountriesincludingTurkey. Amodel, which can truly estimate the future mortality rates with a 95% confidence interval, is built. During the study, a general model which includes the dependent variable; logit function of death rate, and the independent variables; general tendency of the mortality evolution independent of age, the mortality steepness, additional effects of childhood, youth and old age, has been used. The model has two dimensions: the age of the target unit and the year of interest. Generalized linear model estimation, which is a deterministic approach, is used in order to estimate the parameters of the general model. Weighted Least Square Method with a stochastic term, which can be considered as both deterministic and stochastic approach, is used to project future values of one parameter. On the other hand, Random Walk with Drift is used on other parameters. Monte Carlo Simulation is used in order to make the model more stable. The study is applied on Hong Kong [6], USA [13], and Turkish Mortality Rates [23]. Eventually, the life insurance companies and Turkish Social Security Institution can take the advantage of future mortality projections obtained by using the model in this study for the calculation of premiums.