Misspecifying Experimentation
Stuart School of Business research presentation by: Assistant Professor of Business Analytics and Strategy Stanton Hudja
Misspecifying Experimentation
- Assistant Professor of Business Analytics and Strategy Stanton Hudja
Abstract:
This paper analyzes various factors that may be related to behavior in bandit problems. In an experiment, we relate behavior in the first decision of a bandit problem to attitudes towards compound objective lotteries, belief updating, risk aversion, cognitive reflection ability, and contingent reasoning. We find a significant relationship between subject behavior in the first decision of a bandit problem and subject behavior in a similarly composed problem of compound lotteries. However, we fail to find any aggregate relationship between behavior in the first decision of a bandit problem and belief updating, risk aversion, cognitive reflection ability, and contingent reasoning.
All Illinois Tech faculty, students, and staff are invited to attend.
The Friday Research Presentations series showcases ongoing academic research projects conducted by Stuart School of Business faculty and students, as well as guest presentations by Illinois Tech colleagues, business professionals, and faculty from other leading business schools.