Do We Reject Too Often? Small Sample Properties of Tests of Rational Expectations Models

14 Pages Posted: 2 Jan 2007 Last revised: 23 Aug 2024

See all articles by N. Gregory Mankiw

N. Gregory Mankiw

Harvard University - Department of Economics; National Bureau of Economic Research (NBER)

Matthew D. Shapiro

University of Michigan at Ann Arbor - Department of Economics; National Bureau of Economic Research (NBER)

Date Written: October 1985

Abstract

We examine the small sample properties of tests of rational expectations models. We show using Monte Carlo experiments that the asymptotic distribution of test statistics can be extremely misleading when the tine series examined are highly autoregressive. In particular, a practitioner relying on the asymptotic distribution will reject true models too frequently. We also show that this problem is especially severe with detrended data. We present correct small sample critical values for our canonical problem.

Suggested Citation

Mankiw, N. Gregory and Shapiro, Matthew D., Do We Reject Too Often? Small Sample Properties of Tests of Rational Expectations Models (October 1985). NBER Working Paper No. t0051, Available at SSRN: https://ssrn.com/abstract=579743

N. Gregory Mankiw (Contact Author)

Harvard University - Department of Economics ( email )

Littauer Center
Room 223
Cambridge, MA 02138
United States
617-495-4301 (Phone)
617-495-7730 (Fax)

National Bureau of Economic Research (NBER)

1050 Massachusetts Avenue
Cambridge, MA 02138
United States

Matthew D. Shapiro

University of Michigan at Ann Arbor - Department of Economics ( email )

and Survey Research Center
611 Tappan Street
Ann Arbor, MI 48109-1220
United States
313-764-5419 (Phone)
313-764-2769 (Fax)

National Bureau of Economic Research (NBER) ( email )

1050 Massachusetts Avenue
Cambridge, MA 02138
United States
313-764-5419 (Phone)
313-764-2769 (Fax)