Short gap versus Long differences in diff-in-diff event studies: part 2
So, in the previous substack I noted that the OLS event studies always use the omitted year as the baseline in calculating all the event study coefficients, and furthermore that all coefficients are simple “four average and three subtraction” calculations with respect to that omitted year as the baseline.
Short gap versus Long differences in diff-in-diff event study: Part 1
This post is about the event study and the new diff-in-diff methods, but it isn’t about heterogenous treatment effects or differential timing. It’s more general than either of those things, which is why I think this post is probably broadly interpretable even if those topics are not ones you have felt the most confident about.
But now I want to present to you an alternative c…
Keep reading with a 7-day free trial
Subscribe to Scott's Mixtape Substack to keep reading this post and get 7 days of free access to the full post archives.