Friday, 02/02/2024 to 02/02/2024, 12:00 pm to 1:30 pm. ARCHIVED EVENT
Coady Wing, Associate Professor, School of Public and Environmental Affairs, Indiana University
In the most basic event study research design, outcomes in two groups are followed over multiple time periods. At some point, one group is exposed to an intervention and the other is not. Under common trends and no-anticipation assumptions, changes in the difference between the two groups recovers the casual effect of the intervention and how that effect changes over time since exposure. In more complicated designs, there are more than two groups and interventions in each group may occur at different calendar times. This session will introduce the basic assumptions and estimation techniques used to analyze event study designs, as well as common strategies for investigating threats to the validity. We will discuss examples from recent applications to help make the concepts concrete, and we will examine some simple examples using Stata and R to implement analysis in practice.