Do the Obese Respond to Adverse Health Events?: A Bayesian and Behavioral Approach

Abstract

Obesity is an epidemic that is particularly prevalent amongst aging Americans. The economic literature concerning this phenomenon is minimal, and largely focuses on its causes. This dissertation examines the issue from a different direction. First, a Bayesian model is developed to determine how aging Americans use obesity related health information, and if their response differs from the non-obese. Secondly, a behavioral model is employed to see if the same health information elicits a behavioral response. This is the first behavioral response model to determine the existence and magnitude of BMI defined behavioral changes after exposure to new personal obesity related risk information. The estimates of the first model suggest that the obese are Bayesian updaters to a certain extent, and that they utilize adverse health information differently than the non-obese. The behavioral estimates imply that the obese adjust their weight after an adverse health diagnosis. Several suggestions for further work and extensions of these models are given.