Academy of Financial Services Annual Meeting 2022

Machine learning, life insurance, COVID-19

The purpose of this study is threefold. The first is to move beyond traditional modeling techniques to identify the determinants of life insurance demand among rural households using advanced modeling approaches. Specifically, this study applies machine learning algorithms to predict the ownership of cash value life insurance in 2019 (before COVID-19) and 2021 (during COVID-19). The second aim is to identify the best predictors of cash value life insurance ownership in 2019 and 2021 using an optimized machine learning model. The third aim is to compare the list of important predictor variables—as identified using an optimized machine learning model—across periods to identify similarities and differences, and in doing so, add to the existing literature that describes the variables most significantly associated with life insurance ownership.