Gray Calhoun

About Me

I've been an Assistant Professor in Iowa State's Economics Department since August, 2009. Before that, I attended UC San Diego for graduate school. I'm studying econometrics and am particularly interested in settings where we are considering using a complicated model that may be estimated imprecisely, as often happens in forecasting, macroeconomics, and finance. See my C.V. for more details.

Teaching Information

I'm teaching Economics 671 in Fall, 2009, the fisrt econometrics class in the Ph.D. sequence. We will cover basic probability and statistics as well as regression. Please see the webpage for the syllabus and information about class times, etc.

Working Papers

Hypothesis Testing in Linear Regression when k/n is Large

This paper derives the asymptotic distribution of the F-test for the significance of linear regression coefficients as both the number of regressors, k, and the number of observations, n, increase together so that their ratio remains positive in the limit. The conventional critical values for this test statistic are too small, and the standard version of the F-test is invalid under this asymptotic theory. This paper provides new critical values that give correctly-sized tests under both this paper's limit theory and also under conventional asymptotic theory that keeps k finite. This paper also presents simulations that indicate the new statistic can perform better in small samples than the conventional test. The statistic is then used to re-examine Olivei and Tenreyro's results from "The Timing of Monetary Policy Shocks"(2007, AER) and Sala-i-Martin's results from "I Just Ran Two Million Regressions" (1997, AER).

Limit Theory for Comparing Overfit Models Out-of-Sample

This paper uses dimension asymptotics to study the rationale for comparing overfit forecasts out-of-sample instead of in-sample. The two models to be compared are nested linear regressions, and the number of predictions used by the larger model, k, increases with the number of observations, T, so that the ratio k/T remains uniformly positive. Under this limit theory, tests that are designed to reject if the larger model is true, such as the usual in-sample Wald and LM tests and also Clark and McCracken's (2001, Journal of Econometrics) and Clark and West's (2006, Journal of Econometrics) out-of-sample statistics, will choose the larger model too often when the smaller model is more accurate. I show that the usual out-of-sample test using Gaussian critical values performs as desired, mistakenly choosing the larger model with probability equal to the test's nominal size, as long as the out-of-sample period is very small relative to the total sample size.

Miscellaneous Documents

An extremely subjective outline of the economics job market

I put together an outline of my thoughts about the economics job market, mostly to help the 2010 UCSD graduating class. It reflects only my own experience and opinion (except for a few comments by others), so it is likely to be misleading and incomplete. But, now that I've typed it up, I see no reason not to put it online. Please let me know if you have additional comments that you think should be added.