MODELING AND FORECASTING OF SOCIOECONOMIC MORTALITY DIFFERENTIALS ACROSS SUBPOPULATIONS IN THAILAND

Abstract

This research aims to investigate the associations between mortality rates and socioeconomic status across various subpopulations in Thailand, with a specific focus on average income as represented by GPP per capita. The analysis utilizes the Lee-Carter model and its extensions for modeling and forecasting mortality rates. Furthermore, it evaluates the effectiveness of these models and applies the findings to relevant scenarios, particularly in the context of life insurance. The results indicate that the augmented common factor model is the most appropriate for modeling and forecasting socioeconomic mortality differentials among both male and female subpopulations. Moreover, socioeconomic status significantly influences mortality rates across various subpopulations in Thailand. The findings reveal that older individuals in high GPP per capita groups, specifically those aged 70 and above for males and 60 and above for females, experience the lowest mortality rates compared to other subgroups. In contrast, certain age ranges within low GPP per capita subpopulations—males approximately 12 to 65 years old and females approximately 13 to 57 years old—exhibit the lowest mortality rates relative to other subgroups. These differential mortality rates significantly affect the variability of life insurance benefits.