Re: [R] about kpss.test()

From: Achim Zeileis <>
Date: Thu 10 Mar 2005 - 05:46:43 EST

On Wed, 9 Mar 2005 13:31:16 -0500 (EST) Weiguang Shi wrote:

> Hi All,
> First of all, could you tell me what the "KPSS Level"
> in the output of the test means?

It means that you tested for level stationarity and gives you the test statistic.

> I have a series, x, of periodic data and tried
> kpss.test() on it to verify its stationarity. The

Note that stationarity is the null hypothesis and not the alternative of the KPSS test.

> tests
> gave me the p-value above 0.1. Since the null
> hypothesis N0 is that the series _is_ stationary, this
> means that I cannot reject N0. But the series does
> look
> periodic!
> So does all this say stationarity and periodicity can
> co-exist?

Please check the references of kpss.test or any book on economectric time series analysis for which alternatives the KPSS test was designed. Hint: it's not designed for detecting cyclical variations. Z  

> Thanks,
> Weiguang
> ______________________________________________
> mailing list
> PLEASE do read the posting guide!
> mailing list PLEASE do read the posting guide! Received on Thu Mar 10 05:50:31 2005

This archive was generated by hypermail 2.1.8 : Fri 03 Mar 2006 - 03:30:42 EST