Very interesting analysis! I didn’t realize clustered standard errors weren’t in use pre-2000. Is there are reason you published this result here rather than as a letter in the AER?
I can do both—economics has a long tradition of accepting circulation of “working papers” or “preprints” without jeopardizing publication in journals. In fact, Esther suggested I hold off submitting a comment to the AER until she steps down as editor in a few weeks.
Actually, if clustering was not as common in 2001 as now, it was not rare. She clustered standard errors in the other two chapters in her thesis. My future colleague Mead Over coauthored a program in 1996 for clustering standard errors in instrumental variables regressions in Stata.
Nice. Good luck in getting it accepted! Methinks that if she clustered in the other two chapters, but not this one, she actually did try clustering and didn’t like what it showed… Not that I’m judging; we’ve all done it. I remember hearing a rumor in grad school that there was some time period in the 90s when there was a bug in stata, and basically all the standard errors are incorrect from that time.
I wouldn’t be be too confident. Much more honest explanations exist eg they learnt about the technique later in their PhD and didn’t go back to redo earlier analyses.
Seems doubtful as she published it in the American Economic Review after she’d completed her thesis. I’m surprised the reviewers didn’t require that clustered standard errors be shown, but I can only guess it wasn’t common back then.
Very interesting analysis! I didn’t realize clustered standard errors weren’t in use pre-2000. Is there are reason you published this result here rather than as a letter in the AER?
I can do both—economics has a long tradition of accepting circulation of “working papers” or “preprints” without jeopardizing publication in journals. In fact, Esther suggested I hold off submitting a comment to the AER until she steps down as editor in a few weeks.
Actually, if clustering was not as common in 2001 as now, it was not rare. She clustered standard errors in the other two chapters in her thesis. My future colleague Mead Over coauthored a program in 1996 for clustering standard errors in instrumental variables regressions in Stata.
Nice. Good luck in getting it accepted! Methinks that if she clustered in the other two chapters, but not this one, she actually did try clustering and didn’t like what it showed… Not that I’m judging; we’ve all done it. I remember hearing a rumor in grad school that there was some time period in the 90s when there was a bug in stata, and basically all the standard errors are incorrect from that time.
I wouldn’t be be too confident. Much more honest explanations exist eg they learnt about the technique later in their PhD and didn’t go back to redo earlier analyses.
Seems doubtful as she published it in the American Economic Review after she’d completed her thesis. I’m surprised the reviewers didn’t require that clustered standard errors be shown, but I can only guess it wasn’t common back then.