Dear Ping Zheng,
ipshin, levinlin and madfuller successfully work with panel data. The error messages you report clearly show that ipshin, madfuller and levinlin are not installed in your Stata version.
Type "ssc install levinlin" and so on to install these ado.
FOr the second question, dfuller doesn't work with panel data. This problem could be overcome installing "dfuller2" or "pperron2" a recent ado that let you to infer for unit root using single time series of your panel.
The command is
dfuller2/pperron2 variable if idcode==x, lag() tr nocost ecc.
The command doesn't display the unit root test for each series of the panel. You must specify the cross section to analyse.
dfuller2 and pperron2, together dfgls2 are available in stata as "PANELUNIT" routine (ssc install PANELUNIT)
Best,
Andrea Sisto
Phd Student
University of Eastern Piedmont (Italy)
> The help for -dfuller- explains that just one
> variable should be specified. I'm no expert
> but it appears that -dfuller- cannot be applied
> to panel data.
>
> As for your report that -levinlin- and -ipshin-
> "didn't work", it appears that you have not yet
> installed these programs, which are not part of
> official Stata.
>
> I suggest a thorough study of the Statalist
> FAQ and of its recommendations. In particular,
> the -ssc- command is useful for installing many
> user-written commands.
>
> Nick
> [email protected]
>
> Ping Zheng
>
> > By following the suggestions from the people, tsset has been done
> > firstly, then I've tried several ways to do unit root test but
> > unsuccessful (please see below). what is the commands for
> > dfuller test?
> > I've also tried "levinlin", "ipshin", but didn't work either.
> >
> > tsset id year, yearly
> > panel variable: id, 1 to 28
> > time variable: year, 1984 to 2002
> >
>
> > . dfuller
> > varlist required
> > r(100);
> >
> > . dfuller lfdi lex lim
> > too many variables specified
> > r(103);
> >
> > . dfuller lfdi
> > sample may not include multiple panels
> > r(459);
> >
> > . dfuller lfdi lex
> > too many variables specified
> > r(103);
>
> > . levinlin
> > unrecognized command: levinlin
> > r(199);
> >
> > . ipshin
> > unrecognized command: ipshin
> > r(199);
> >
> > . madfuller
> > unrecognized command: madfuller
> > r(199);
>
> *
> * 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/
>
*
* 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/
