Friday, January 30, 2009

The best paper I read this month

It's by Jesús Fernández-Villaverde and it's titled "The Econometrics of DSGE Models" ( ungated version here).

It lays out the evolution of DSGE models, gives an excellent and readable account of Bayesian estimation, and discusses some of the major challenges facing the field. It's also witty and a pleasure to read.

Here is a sample:

"the likelihood of DSGE models is, as I have just mentioned,a highly dimensional object, with a dozen or so parameters in the simplest cases to close to a hundred in some of the richest models in the literature. Any search in a high dimensional function is fraught with peril. More pointedly, likelihoods of DSGE models are full of local maxima and minima and of nearly flat surfaces. This is due both to the sparsity of the data (quarterly data do not give us the luxury of many observations that micro panels provide) and to the flexibility of DSGE models in generating similar behavior with relatively different combination of parameter values (every time you see a sensitivity analysis claiming that the results of the paper are robust to changes in parameter values, think about flat likelihoods)."

and another:

"A compelling proof of how unnatural it is to think in frequentist terms is to teach introductory statistics. Nearly all students will interpret con…fidence intervals at …first as a probability interval. Only the repeated insistence of the instructor will make a disappointingly small minority of students understand the difference between the two and provide the right interpretation. The rest of the students, of course, would simply memorize the answer for the test in the same way they would memorize a sentence in Aramaic if such a worthless accomplishment were useful to get a passing grade. Neither policy makers nor undergraduate students are silly (they are ignorant, but that is a very different sin); they just think in ways that are more natural to humans. Frequentist statements are beautiful but inconsequential."

People, I am gonna have to go with BOTH "Yikes" AND "Amen". It's well worth reading in its entirety.

PS. The second best paper I read this month is here (forthcoming in the JME but been around for a while).

1 comment:

Anonymous said...

I still find it difficult to convince others by presenting Bayesian results for cross-section data.

For example, to present an OLS regression via Bayesian estimation makes them suspicious that the wool is being pulled over their eyes, rather than it just being a philosophical difference in what data can tell us.

Hopefully more papers like this will illustrate the point, and make it easier to use Bayesian methods to present results.