Monday, October 22, 2007

The Problem With Economics

Read this.

Russ Roberts does it again. It is the best thing I have read in a while.

1 comment:

Stephen said...

Did you get a chance to listen to the podcast that the post was about? I am not sure that he gives Ayers a fair shake. I think Ayers was talking about how statistics can be a great substitute for human experts when real data exists. He talked a lot about Capital One and how they did randomized studies with different mailings. Ayers point was that if you have really solid data then you don't need to rely upon the word of the statistician that they aren't cooking the books. Ayers 2nd point was that econmetrics is a better tool for business than it ever will be for policy or academics. I agree with Russ, that most econometrics work about big policy questions is unconvincing, but that wasn't Ayers point.