Congratulations to Amy Mahler of @usc_econ for defending her thesis today. She will be joining Rand this summer and they are very lucky to have hired her. Our success at USC Economics is due to selection effects and treatment effects!!
Bringing informative priors into your experiment: Bayesian impact evaluation - a blogpost on the paper I'm most excited about at the moment. Joint with @leoiacovone & @economeager https://t.co/Fsfs5w7jek
When you use xtreg, fe in Stata, the last line of output is a test of whether all unit-specific intercepts are the same. It's reported as an F statistic, with N - 1 restrictions being tested. This test is often used as a way to test the pooled OLS analysis against unit FEs.
Hey @KhoaVuUmn this is for you and your kids. I told my daughter this song was ripe for a parody. The astronomers set a high bar, can #EconTwitter top this? https://t.co/GiVWepPlpU
After I graduated I took a class on literature by immigrants of color and I think it will be more helpful for my economics work than real analysis was. Economics is about people and that class taught me a lot about people. Real analysis just made me deeply troubled.
Drinking at least 5 glasses of wine per week also reduces the risk of making causal claims based on an observational study with a selection on observable design.
This morning, I was asked by a young scholar, "How do write for impact?" I shared a process that I've developed over time that is pretty simple, but it's been effective for me. 🧵⬇️⬇️⬇️
Are you overwhelmed with the recent reviews on the new difference-in-differences methods? In this paper, I discuss a new method to synthesize this growing literature.
In the last week I've started to receive inquiries from people running tree planting programs wanting my help. I am suggesting that they shut down their programs. Here I will explain why:
I don't know if folks have seen this, but @AsjadNaqvi did us a huge favor of putting all Stata commands for different DiD methods together here along with other resources.