Relationship of Health and Economic Status with Stock Market Participation among Older Adults : Insights from Machine Learning Approach

This study investigates the relationship between health, wealth, and stock market participation among older adults using machine learning techniques, with a focus on Accumulated Local Effect （ALE） plots within an Artificial Neural Network （ANN） framework. Unlike traditional econometric models, which assume linearity and homogeneity in predictor relationships, our approach captures complex non-linear interactions and heterogeneous associations. The analysis reveals that higher net worth and income significantly increase the likelihood of stock market participation, with the effect being more pronounced among individuals in better health. Conversely, those in poorer health are less likely to invest, likely due to financial constraints or risk aversion. Through feature importance analysis and ALE plots, we offer a detailed understanding of how financial and health variables interact to shape investment behavior. These findings offer valuable insights for policymakers and financial advisors aiming to increase stock market participation among diverse population segments. Keywords : Stock market participation, older adults, health status correlation, economic status, artificial neural networks （ANN）