Although both the Johansen (1991, 1994) trace test and Bierens (1997a,b) nonparametric lambda-min test for cointegration have good size properties in Monte Carlo studies by Hubrich, Lutkepohl, and Saikkonen (2001) and Boswijk, Lucas, and Taylor (2000), the Bierens test has very low power. In contrast, Bierens reports good power for his procedure. Meanwhile, Hubrich et al. and Boswijk et al. do not include Bierens' companion method for estimating the number of cointegrating vectors, nor do they investigate the effect of serial correlation on Bierens' test. In the present paper, inclusion of the estimation step does not significantly degrade size of the Bierens procedure, even with serial correlation, but power is not improved. Serial correlation does degrade the size of the Johansen test, but it remains superior. Analysis of Bierens' (1997b) Monte Carlo results suggests that their indication of high power reflects the test's lack of scale invariance.
Download Info
To download:
If you experience problems downloading a file, check if you have the
proper application to
view it first. Information about this may be contained
in the File-Format links below. In case of further problems read
the IDEAS help
page. Note that these files are not on the IDEAS
site. Please be patient as the files may be large.
Publisher Info
Article provided by Economics Bulletin in its journal Economics Bulletin.
Find related papers by JEL classification: C3 - Mathematical and Quantitative Methods - - Multiple or Simultaneous Equation Models; Multiple Variables
References listed on IDEAS Please report citation or reference errors to , or , if you are the registered author of the cited work, log in to your RePEc Author Service profile, click on "citations" and make appropriate adjustments.: