Joshua Angrist on Econometrics and Causation
EconTalk
00:00
The Importance of Sensitivity Analysis in Empirical Practice
Lemur had a huge impact on me and I think on empirical work all to the good that he is complaining about the kind of arbitrariness of what I report. He was proposing a fairly conventional, I thought, Bayesian approach where you would state your priors and you would then show how that maps. But he also had the idea that we should show many variations. A core concept in my work with, in my writing with Steve and the research methods that I think are most effective is the notion of design.
Play episode from 20:14
Transcript
