Message-ID: <20050503114636.6c5da3e8.Achim.Zeileis@wu-wien.ac.at>
Date: 2005-05-03T09:46:36Z
From: Achim Zeileis
Subject: Trying to understand kpss.test() in tseries package
In-Reply-To: <7E4C06F49D6FEB49BE4B60E5FC92ED7A01DA7E7A@pnlmse35.pnl.gov>
On Mon, 02 May 2005 14:43:02 -0700 Waichler, Scott R wrote:
>
> I'm trying to understand how to use kpss.test() properly. If I have a
> level stationary series like rnorm() in the help page, shouldn't I get
> a small p-value with the null hypothesis set to "Trend"?
No, every level stationary series is also trend stationary (with a zero
trend). Hence, the test statistic is even smaller for null = "Trend"
because first the trend is removed (which is not exactly zero due to
random variation).
Z
> The (condensed)
> output from kpss.test() for the two possible null hypotheses is given
> below. I don't see any significant difference between these results.
>
>
> > x <- rnorm(1000) # is level stationary
> > kpss.test(x, null="Level")
> KPSS Test for Level Stationarity
> KPSS Level = 0.0638, Truncation lag parameter = 7, p-value = 0.1
> Warning: p-value greater than printed p-value
>
> > kpss.test(x, null="Trend")
> KPSS Test for Trend Stationarity
> KPSS Trend = 0.0275, Truncation lag parameter = 7, p-value = 0.1
> Warning: p-value greater than printed p-value
>
> I can't get the original reference easily.
>
> Scott Waichler
> Pacific Northwest National Laboratory
> scott.waichler at pnl.gov
>
> ______________________________________________
> 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
>