Author(s): Brendon McConnell
Looking for co-authors: No { message_status }
With a dependent variable in logs, the difference-in-differences (DD) term does not capture the outcome difference between treated and untreated groups over time. Rather it reflects an approximation of the proportional difference in growth rates across groups. As I show with both simulations and three published case studies, if the baseline outcome distributions are sufficiently different across groups, the DD parameter for a log-specification can be different in sign to that of a levels specification. I provide a condition, based on (i) the aggregate time effect, and (ii) the difference in relative baseline outcome means, for when the sign-switch will occur.
Published: 2024-04-09 16:06:46 PT
Stage: Working Paper
Fields: Econometrics
Research Group(s): Playground
Referee Reports: No referee reports yet. (Add your report)
Versions: v1 (04/09/2024)