Gender Differences in U.S. Life Expectancy Forecasts: A State-Level Analysis
Open Access
Author:
Morrison, Justin
Area of Honors:
Actuarial Science
Degree:
Bachelor of Science
Document Type:
Thesis
Thesis Supervisors:
Nan Zhu, Thesis Supervisor Amanda W Hammell, Thesis Honors Advisor
Keywords:
Life Expectancy Mortality United States Time Series Double-Gap Life Expectancy Model ARIMA Gender Actuarial Science Insurance Demography
Abstract:
The level and trend of life expectancy are key forces of concern to both public policymakers and insurance companies; it is an indicator of the quality of life of the general population or the financial obligation to insurance policyholders. Understanding historical life expectancy trends in the U.S. already has significant literature and research. Many studies have examined some of the demographic and socioeconomic differences in life expectancy. However, forecasting the direction of mortality improvement at the state level is an area of research with untapped potential. In this thesis, the Double-Gap Life Expectancy model, a combination of simple linear regression and ARIMA time series models, is used to forecast life expectancy for gender-specific and aged-based sub-populations for the six most populous states in the U.S. This paper quantifies differences in life expectancy across these representative states and explores qualitative reasons for these differences using a correlation-based approach.
