Forecasting Population Mortality Rates Using Generalized Lee-Carter Model

Abstract

The demographic process cannot be inseparable from the mortality rate. The appropriate models for forecasting mortality rates are essential in assisting governments, companies, and other agencies in formulating policies or making decisions. As one of the countries with the highest death rate, Japan is influenced by several factors. This research uses the Generalized Lee-Carter Model), which is one of the developments of the Lee-Carter (LC) model. The Lee-Carter model was prevalent by Lee and Carter (1995) as an alternative that is suspected to predict the mortality rate of an area. The first step in this research is to formulate the Generalized Lee-Carter function. Through the function formula, the estimator value of the Generalized Lee-Carter model will be searched in the second stage. And the third stage, through the Generalized Lee-Carter model, will find the RMSE value and then use it in the fourth stage, namely forecasting the future period using ARIMA. The data in this study is facilitated through www.mortality.org, which is one of the Japanese population data. The result of the study showed that the RMSE value for females was 0.01670 and 0.016292 for males. So, it concluded that the Generalized Lee-Carter Model is great for forecasting mortality rates.