ADAPTED PRESTON’S CURVE: A PROXY METHOD FOR LONGEVITY RISK ANALYSIS ON INDONESIAN PENSION PLAN

Abstract

Future lifetime will increase as the standard of living and the health insurance system develop. This increase will have an effect on financial contracts' actuarial present values, particularly the liabilities of pension funds. Longer-lived retirees will have more financial obligations to the pension plan in the future. Preston established a link between GDP and life expectancy at birth, which served as the inspiration for this paper's concept. We strive to advance Preston's work on longevity analysis, particularly how to create a proxy approach for capturing the dynamic of the mortality model with other data. In this case, we utilize the Lee-Carter model to capture the long-term dynamics of the mortality rate, and our GDP-related measure will be based on the model's parameters. We use the Human MortD data to gather the longevity parameter’s estimate and fit the relationships using linear, local linear, and kernel regressions. Since the long-term goal of this study is longevity risk management in Indonesia, hence the model's applicability is assessed by how closely it resembles Indonesia's mortality models. We discovered that the linear model, which has an RMSE of 2.19234, has the lowest RMSE, then we conclude that the long-term relationships between longevity parameters and GDP can be explained by the linear model.