Forecasting Indonesian mortality rates using by Lee-Carter model and Regression Linear model

Abstract

Premium price in life insurance is determined by the mortality rate of the country concerned. In this research, we construct Indonesia mortality table using the Lee-Carter model and parameters in the model are estimated by Least Square method and Newton Raphson method. After that, the parameter that depends on time will be forecasted by Linear Regression model. The accuracy of estimated the value of parameters is the most important and it can be measure with MAPE (Mean Absolute Percentage Error), with using Least Square method and Newton Raphson method have the smallest MAPE value for all age is not more than 10%. Thus, those method can be used. Forecasting Indonesia mortality table using Linear Regression model can be done because all classical assumptions in Linear Regression model are ful-filled and it has MAPE value is 11.97%. The final result of this research is Indonesia mortality table for some future periods.