Andrew Gelman linked to an interesting and short essay by the newly minted Economics Nobelist Christopher Sims. Sims won the Nobel for his work on macroeconometrics – specifically, vector autoregressions that were a big deal in the late 70s and 80s (to present) in helping to parse causality in macroeconomics. The linked essay is a critique of Angrist and Pischke, two prominent microeconometricians* who promote specific recent innovations. Here’s Sims best takeaway line:
Natural experiments, difference-in-difference, and regression discontinuity design are good ideas. They have not taken the con out of econometrics — in fact, as with any popular econometric technique, they in some cases have become the vector by which “con” is introduced into applied studies. (p. 8 )
An interesting (if somewhat dense) read, if you’re interested in the practical consequences of innovations in statistical techniques, and some of the problems that innovation can produce.
*Whose book Mostly Harmless Econometrics is a pretty fun read, given that it’s a book about econometrics.