Comparison of Brass Logit and Kostaki Accuracy in Life Table Reconstruction for Insurance Premiums

Despite the widespread use of interpolation methods in mortality studies, limited research has conducted a comprehensive comparison of their statistical accuracy and financial implications in the Indonesian context. This study addresses the limited availability of complete mortality data in Indonesia, where mortality information is commonly presented in abridged life tables with five-year age intervals, creating challenges for accurate premium determination. The objective of this research is to compare the performance of the Brass Logit and Kostaki methods in reconstructing complete life tables and to evaluate their impact on term life insurance premium estimation. A quantitative comparative approach is employed using secondary data from the Coale Demeny West Model, Indonesia’s life expectancy published by Statistics Indonesia, and the Indonesian Mortality Table IV as the benchmark. The analysis involves model level selection, interpolation, accuracy evaluation using Root Mean Square Error, and application to premium calculations. The results show that the Brass Logit method provides higher statistical accuracy, while the Kostaki method produces premium estimates closer to the benchmark. These findings highlight a trade-off between statistical precision and financial relevance, offering practical guidance for method selection and supporting more robust actuarial decision making in the insurance industry