Modelling USA Age-Cohort Mortality: A Comparison of Multi-Factor Affine Mortality Models

Affine mortality models are well suited for theoretical and practical application in pricing risk management of risk. They produce consistent, closed-form stochastic survival curves allowing the efficient valuation mortality-linked claims. We model USA age-cohort data using five multi-factor affine models. focus on three-factor compare four Gaussian along with a based Cox–Ingersoll–Ross (CIR) process, Gamma-distributed rates. assess Arbitrage-Free Nelson–Siegel (AFNS) model, which incorporates level, slope curvature factors, canonical factor both without correlations dynamics. show that data, probability negative rates is small. Models estimated discrete time versions capturing variability cohort curves. Poisson variation included estimation Kalman filter through measurement equation. consider incorporating dependence to capture effects age-dependence The analysis demonstrates independent-factor AFNS performs compared other explaining forecasting data.