PP test uses an estimate of the spectral density at frequency 0. For
nonstationary processes this is an ill-posed estimation problem. This is
another issue with the PP test. As far as I know, the ADF test does not have
this problem.
Best wishes,
Mehmet
Quoting Adrian Trapletti <adrian.trapletti at lmttrading.com>:
> > Date: Mon, 22 Apr 2002 07:21:41 +0000
> > From: "sonchawan tamkaew" <psu17772 at hotmail.com>
> > Subject: [R] PP.test
> >
> > Hello there,
> >
> > I would like to know whether PP.test from library(ts) is appropriate
to
> test
> > unit root in financial data (daily)?
> >
> > Thank you for your help.
> >
> > Sonchawan
> >
>
> In principle yes. However, it is by now a well know fact that the PP test
has
> severe size distortions for certain processes (e.g. integrated MA(1)). A
more
> robust alternative in some situations is the Augmented-Dickey-Fuller test
> which is provided by tseries (see adf.test).
>
> best
> Adrian
>
>
>
>
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
.-> r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
> Send "info", "help", or "[un]subscribe"
> (in the "body", not the subject !) To: r-help-request at
stat.math.ethz.ch
>
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._>
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !) To: r-help-request at
stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._