st: Re: KPSS test in STATA

[Kit Baum <baum@bc.edu>]

Dear Jee-Kwang

The kpss command returns the statistics as individual scalars, not as a 
matrix:

. webuse wpi1

. whelp kpss

. kpss wpi

KPSS test for wpi

Maxlag = 4 chosen by Schwert criterion
Autocovariances weighted by Bartlett kernel

Critical values for H0: wpi is trend stationary

10%: 0.119  5% : 0.146  2.5%: 0.176  1% : 0.216

Lag order    Test statistic
    0           1.96
    1           .992
    2           .669
    3           .507
    4           .411

. return list

scalars:
             r(kpss4) =  .4109284269005028
             r(kpss3) =  .5073039368808655
             r(kpss2) =  .6685610195922723
             r(kpss1) =  .9915345342595147
             r(kpss0) =  1.958072535544316
                 r(N) =  124

.
If you specify the maxlag option, you will know how many of these there 
are. If you want to put them into a matrix, it is easy enough to do so:


. mat kpres = J(5,1,0)

. forv i=1/5 {
  2. mat kpres[`i',1] = `r(kpss`=`i'-1')'
  3. }

. mat list kpres

kpres[5,1]
           c1
r1  1.9580725
r2  .99153453
r3  .66856102
r4  .50730394
r5  .41092843


Kit Baum, Boston College Economics
http://ideas.repec.org/e/pba1.html

On Apr 11, 2005, at 12:02 AM, jp870@columbia.edu wrote:


> Dear Professor Baum,
>
> I am using your KPSS test command. Thanks for contributing the
> command to STATA community.
> I have a question in getting t-statistics after running the KPSS
> test.
> The "dfgls" command save the results as "r(results)" so that I can
> store it as a matrix of a new name.
> How can I get the results(test-statistics) matrix after KPSS?
> I will appreciate your response deeply.
>
> Bests
>
> Jee-Kwang Park
> Department of Political Science
> Columbia University
*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/