Do We Reject Too Often? Small Sample Bias in Tests of Rational Expectations

We examine the small sample properties of tests of rational expectations models. We show using Monte Carlo experiments that these tests can be extremely biased toward rejection for sample sizes typical in applied research. These biases are important when the time series examined are highly autoregressive. We also show that these tests are even more biased with detrended data. We present correct small sample critical values for our canonical problem.

MoreLess

Year of publication:	1985-04
Authors:	Mankiw, N. Gregory ; Shapiro, Matthew D.
Institutions:	Cowles Foundation for Research in Economics, Yale University
Subject:	Rational expectations \| non-stationary time series \| detrending \| small sample bias

More details

Extent:	application/pdf
Series:	Cowles Foundation Discussion Papers.
Type of publication:	Book / Working Paper
Language:	English
Notes:	CFP 637. Published in Economics Letters (1986), 20: 139-145 The price is None Number 743 20 pages
Source:	RePEc - Research Papers in Economics

Persistent link: https://www.econbiz.de/10004990678