Determinants and Predictions of Risks of Diseases in Mid Ages: Logistic Regression Models versus Deep Neural Network Models

Abstract

The prediction of risks of various diseases and identification of factors that influence these risks are important for public policies and disease diagnosis in healthcare. The biomedical literature suggests that much of an individual’s later life health outcomes is programmed at early stages of life. The programming is strongly modulated throughout life by epigenetic inputs such as psychological, financial, social or chemical stress, diets, smoking, substance use, and exercising, with stronger effects imparted in early stages of life. Traditionally estimation of the effect of these factors on risks of diseases is carried out in the statistical logistic regression framework. More recently, the deep neural network framework has shown superior predictive performance in other fields. Using the confusion matrix and other indicators, the paper compares the effectiveness of these two approaches in predicting and identifying the influential observable characteristics that are strongly associated with these risks. The paper uses the Health and Retirement Studies (HRS) dataset.