The Wald test for linear restrictions on cointegrating vectors
is compared in finite samples using the Monte Carlo method.
The Wald test is calculated within the vector error-correction
based estimation methods of Bewley, Orden, Yang, and Fisher
(1994, Journal of Econometrics 64, 3–27) and
of Johansen (1991, Econometrica 59, 1551–1580),
the canonical cointegration method of Park (1992,
Econometrica 60, 119–143), the dynamic ordinary
least squares method of Phillips and Loretan (1991, Review
of Economic Studies 58, 407–436), Saikkonen (1991,
Econometric Theory 7, 1–21), and Stock and Watson
(1993, Econometrica 61, 783–820), the fully modified
ordinary least squares method of Phillips and Hansen (1990,
Review of Economic Studies 57, 99–125), and the
band spectral techniques of Phillips (1991, in W. Barnett, J.
Powell, & G. E. Tauchen (eds.), Nonparametric and
Semiparametric Methods in Economics and Statistics, pp.
413–435). The Wald test performance is also compared to
that of the likelihood ratio test suggested by Johansen and
Juselius (1990, Oxford Bulletin of Economics and
Statistics 52, 169–210) and to a Bartlett correction
of that test as proposed by Johansen (1998, A Small Sample Test
for Tests of Hypotheses on Cointegrating Vectors, European