Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, statalist.org is already up and running.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: A query regarding Dickey-Fuller and Kwiatkowski-Phillips-Schmidt-Shin test results


From   GUCBILMEZ Ufuk <U.Gucbilmez@ed.ac.uk>
To   "'statalist@hsphsun2.harvard.edu'" <statalist@hsphsun2.harvard.edu>
Subject   st: A query regarding Dickey-Fuller and Kwiatkowski-Phillips-Schmidt-Shin test results
Date   Mon, 19 Mar 2012 14:58:33 +0000

Dear Statalist,

I am working with monthly time series data, and am trying to test for the (level) stationarity of the series. The Dickey-Fuller test rejects the existence of a unit root at 1% (see below). Alternative unit-root tests (Phillips-Perron and DF-GLS) yield the same conclusion.

As far as I understand, the Dickey-Fuller test has low power. So, I complement it with a Kwiatkowski-Phillips-Schmidt-Shin stationarity test on the same series (see also below). The KPSS test fails to reject stationarity at 1% for lags higher than 3, but rejects it at the same significance level for up to 3 lags.

My question is whether I can safely conclude that the series is stationary based on the Dickey-Fuller test stat. & the KPSS test stat. with more than 3 lags, or should I be concerned with the rejection of null when the KPSS test stat. has up to 3 lags? 

Many thanks for your advice,
Ufuk

---------------------------------------
Dr Ufuk Gucbilmez
University of Edinburgh


. dfuller VolumeGm

Dickey-Fuller	test for unit	root                   Number of obs   =       623

		---------- Interpolated Dickey-Fuller ---------
	Test		1% Critical       5% Critical      10% Critical	
	Statistic	Value             Value             Value		
											
Z(t)	-6.806	-3.430            -2.860            -2.570	


. kpss VolumeGm, notrend
 
KPSS test for VolumeGm
 
Maxlag = 18 chosen by Schwert criterion
Autocovariances weighted by Bartlett kernel
 
Critical values for H0: VolumeGm is level stationary
 
10%: 0.347  5% : 0.463  2.5%: 0.574  1% : 0.739
 
Lag order    Test statistic
    0           3.06
    1           1.64
    2           1.14
    3           .882
    4           .721
    5           .613
    6           .536
    7           .478
    8           .432
    9           .396
   10           .367
   11           .343
   12           .322
   13           .305
   14            .29
   15           .277
   16           .265
   17           .255
   18           .247 


-- 
The University of Edinburgh is a charitable body, registered in
Scotland, with registration number SC005336.


*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/


© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index