Consumer Choice Modeling in Microsimulation

• Activity Number: 478
• Type: Invited
• Date/Time: Wednesday, August 6, 2014 : 10:30 AM to 12:20 PM
• Sponsor: Section on Bayesian Statistical Science
• Author(s): Laura Hatfield*+ and Thomas McGuire and Michael Chernew
• Companies: Harvard Medical School and Harvard Medical School and Harvard Medical School
• Keywords: microsimulation ; health care policy ; Bayesian ; consumer choice

Abstract

We expand an established dynamic microsimulation of retirees in the United States, adding health care spending and supplemental health insurance to health status, wealth, and income. To enable policy experiments, we incorporate income effects of premiums and cost-sharing, trends in benefit design, medical technology growth, and supplemental coverage. In this talk, I will focus on two systems for transitions in supplemental coverage. We wish to build a system in which individuals sort realistically according to their personal characteristics. A key challenge in a complex model is calibrating the many parameters to observed data. I will highlight Bayesian contributions to calibration and compare the two models in terms of usefulness, realism, and fit.