[R] Question regarding Johansen's cointegration testing

From: <adschai_at_optonline.net>
Date: Sat, 02 Jun 2007 03:49:59 +0000 (GMT)


I have a couple of questions about johansen's test, in general:

  1. I was able to obtain error correction term (ect) from cajorls$rlm$model properly. According the my ca.jo object on 2-variate series, the test suggests that the integration rank is 1. Which means that my ect should be stationary. However, I did test stationariy on ect and it shows non-stationarity and my acf still shows high and long autocorrelation. How should I interpret the result from Johansen then?
  2. In some cases for 2-variate series, I found that Johansen testing shows that my test statistics are lower than the 5% for both r <= 0 and r <= 1. That means I have 1 or 0 cointegration relationship? How should I interpret this result?

Thank you,

        [[alternative HTML version deleted]]

R-help_at_stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. Received on Sat 02 Jun 2007 - 03:59:28 GMT

Archive maintained by Robert King, hosted by the discipline of statistics at the University of Newcastle, Australia.
Archive generated by hypermail 2.2.0, at Sat 02 Jun 2007 - 04:31:24 GMT.

Mailing list information is available at https://stat.ethz.ch/mailman/listinfo/r-help. Please read the posting guide before posting to the list.