Monte Carlo evidence on the power of the Durbin-Watson test against nonsense relationships
It is a well-known fact that, in linear regressions involving the levels of integeated processes spuriously related, the Durbin-Watson statistic converges in probability to zero. This, in turn, implies that this statistic can provide an useful testing procedure against the presence of nonsense relationships. Marmol (1997) extends this result to the case of spurious regressions among nonstationary fractionally integrated processes. The aim of this paper is to provide a theoretical overview of these asymptotic results as well as Monte Carlo evidence on the behavior of the Durbin-Watson test in small samples.
|Date of creation:||Mar 1997|
|Contact details of provider:|| Web page: http://portal.uc3m.es/portal/page/portal/dpto_estadistica|
When requesting a correction, please mention this item's handle: RePEc:cte:wsrepe:6209. See general information about how to correct material in RePEc.
For technical questions regarding this item, or to correct its authors, title, abstract, bibliographic or download information, contact: (Ana Poveda)
If references are entirely missing, you can add them using this form.