Selection of a Stochastic Mortality Model for the Male and Female Population in Iran

Forecsting the death rate accurately is of paramount importance. The purpose of this research is to select the best stochastic mortality model in Iran based on a comparison between Lee-Carter models and the age-at-death distribution model (maximum entropy) for the male and female populations separately. It is worth mentioning that one of the strengths of modeling using age-at-death distribution is that it effectively represents the age-specific mortality pattern in Iran and answers the question of whether the mortality pattern in Iran is stable. Considering any model that reports the lowest error based on evaluation criteria, the life table for the year 1400 in Iran has been introduced and compared with the mortality tables TD88-90 in France and ILT1400 in Iran. In this regard, real mortality data in Iran for the years 2011-2019 for ages 0-100 years have been used. Ultimately, it was concluded that the Lee-Carter model performed better for the male population and the Currie model for the female population in Iran. It was also concluded that the existing tables in Iran report a higher mortality rate, and the life table presented in this article may be suitable for pricing insurance products and for more accurate calculation of actuarial indicators, such as the probability of death.