All told, Mostly Harmless Econometrics: An Empiricist's Companion is neither "mostly harmless" nor an "empiricist's companion." Rather, it's a companion for a highly-specialized group of applied non-structural micro-econometricians hoping to estimate causal effects using non-experimental data and largely-static, linear, regression-based methods. It's a novel treatment of that sub-sub-sub-area of applied econometrics, but pretending to be anything more is most definitely harmful, particularly to students, who have no way to recognize the charade as a charade.Disclaimer, I haven't read the book. The quote does summarize feelings I have had in many seminars involving difference in difference in difference regressions with 100 fixed effects and controls. But mostly I post it as a lovely quote.
Sunday, February 19, 2017
Good Review
Frank Diebold, on Mostly Harmless Econometrics:
8 comments:
Comments are welcome. Keep it short, polite, and on topic.
Thanks to a few abusers I am now moderating comments. I welcome thoughtful disagreement. I will block comments with insulting or abusive language. I'm also blocking totally inane comments. Try to make some sense. I am much more likely to allow critical comments if you have the honesty and courage to use your real name.
Candor like this can be really useful for students making big decisions about their research careers.
ReplyDeleteReminds me of the line from the original Alien movie:
ReplyDelete"We better get back because it will be dark soon and they mostly come at night, mostly" - Newt
But I believe the title is actually a reference to the Douglas Adams book:
https://en.wikipedia.org/wiki/Mostly_Harmless
Totally agree. For those interested, I think that this:
ReplyDeletehttps://www.aeaweb.org/articles?id=10.1257/jep.24.2.47
is probably the best Angrist-Pischke bashing paper out there. (and it is one of the funniest(!) econometrics paper I've read as well)
The Grumpy Economist meets the Grumpy Econometrician!
ReplyDeleteMostly Harmless isn’t about everything, but it’s about some really useful things. For those of you who--like me--just aren’t very smart, the book does a nice job of connecting techniques that you see everyday (e.g., dif-dif and fixed effects) to the experimental ideal. It does this with accessible examples from real micro data about stuff we actually care about and, even better, works in lots of references to the Douglas Adams take on “Mostly Harmless”. Angrist and Pischke also have another book, “Mastering Metrics” that is helpful.
If you want to preview Angrist before you commit to using the books, Russ Roberts did a nice Econtalk podcast with him a couple of years ago. (I’ll try and insert the link, but if I can’t, it’s in the Econtalk archives.)
http://www.econtalk.org/archives/2014/12/joshua_angrist.html
Wow. You really should read the book, professor. I'm sure after reading it you'll realize that quote is extremely unfair. The book strives to explain in an intuitive manner the nature of causation, and exposes to the reader some of the implicit challenges in emperical econometrics. I found it to be humble, not charade...
ReplyDeleteYou are likely right. The review is clever, and my frustration at the bad empirical work I often have to sit through does not deny the huge contribution that good micro-empirical work has made.
DeleteI wonder what the same person thinks of Imbens and Rubins new book. One could argue that that book and MHE are siblings.
ReplyDeleteReading these comments I'm starting to see that there is no such thing as an unbiased assessment. But also that the answer to speech is more speech.
ReplyDelete