Michael Pollmann - Duke University

Abstract

This paper develops a method to infer causal effects of treatments on choices, by exploiting relationships between choices and hypothetical evaluations. Under specified conditions, it can recover treatment effects even if the treatment does not vary across observations in the sample. Additional advantages include more comprehensive recovery of heterogeneous treatment effects and potential improvements in precision. These advantages can also be attained in some environments where treatment is assigned endogenously. We provide proof of concept by using the approach to estimate the price responsiveness of the demand for snack foods in the laboratory, and the response of contributions to the availability of matching funds on a microfinance website.

Additional information:

• Speaker: Michael Pollmann
• Time: Thursday, 22.05.2025, 16:00 - 17:00
• Location: Faculty Lounge, Room 0.036
• Further links:
  • Michael Pollmann's webpage¹
• Organizer: Statistics Group
• Contact:
  • Almut Lunkenheimer
  • +49 228 73-9228
  • ifs@uni-bonn.de