A Markovian Aging Process Forecasting Model: Predicting U.S. Mortality (Q7000893)

Forecasting mortality rates and life expectancy has long been a crucial problem in many scientific disciplines, including insurance, economics, and demographics.\N\NThe authors ``provide a novel finite-state Markov model for predicting death rates. The Markovian physiological age, which forms the basis of this model, is represented by the states in the underlying continuous-time Markov chain. This model forecasts mortality rates for the Markovian physiological age, for which mortality rates for calendar ages can be easily computed. A set of data from the U.S. population is used to calibrate the model. The data collection includes individuals aged 30 to 108 and covers the years 1970 to 2019.'' \N\NThe model is trained using data from 1970 to 2014 and then it is tested using data from 2015 to 2019. \N\N``Based on metrics utilized for training and test datasets, the suggested model outperforms the models of both Lee and Carter [\textit{R. D. Lee} and \textit{L. R. Carter}, J. Am. Stat. Assoc. 87, No. 419, 659--675 (1992; Zbl 1351.62186)] and Renshaw and Haberman [\textit{A. E. Renshaw} and \textit{S. Haberman}, Insur. Math. Econ. 38, No. 3, 556--570 (2006; Zbl 1168.91418)].'' \N\NThe model uses fewer model parameters than the comparable models and, unlike the Lee Carter model, is identifiable.