Usually, I release a new podcast every Tuesday, but my producer is not yet finished with the new music and the new montage for season 3’s opener. So this week I’m going to provide a free substack in which I express my personal opinion about why I prefer event study plots a particular way over another way, even though I have a recent paper doing it the way I don’t like. I prefer presenting your coefficients and confidence intervals in “whisker plot” or what I’ll call here the disconnected method to connecting all point estimates (and their confidence intervals) to the nearest neighbor with straight lines, or what I’ll just call the connected method. I have a few reasons for this which I’l review below.
Some examples of both ways
You can find tons of examples of both ways of displaying the coefficients and confidence intervals from event study regressions in fantastic papers. Here’s three examples of the “connected” method, one of which is from a new paper of mine.
And here’s a great…
Keep reading with a 7-day free trial
Subscribe to Scott's Substack to keep reading this post and get 7 days of free access to the full post archives.