OK, so that's seemingly not the explanation.
As reported, the original example in the help file
works fine with me. See below. So, I don't have
any more ideas here.
. use http://fmwww.bc.edu/ec-p/data/hayashi/sheston91.dta,clear
(Penn World Table Mark 5, R. Summers / A. Heston, 1991)
. madfuller rgdppc if country<11, lags(2(2)8)
Multivariate Augmented Dickey-Fuller test for rgdppc
with 26 observations on 10 cross-sectional units
-----------------------------------------------------------
Obs Lags MADF Approx 5% CV
-----------------------------------------------------------
24 2 41.274 31.844
22 4 45.038 34.737
20 6 30.610 38.897
18 8 51.578 45.195
-----------------------------------------------------------
H0: all 10 timeseries in the panel are I(1) processes
. madfuller D.rgdppc if country<11, lags(4)
Multivariate Augmented Dickey-Fuller test for D.rgdppc
with 25 observations on 10 cross-sectional units
-----------------------------------------------------------
Obs Lags MADF Approx 5% CV
-----------------------------------------------------------
21 4 82.132 36.616
-----------------------------------------------------------
H0: all 10 timeseries in the panel are I(1) processes
Nick
[email protected]
www
> I have update all the ado file and executable file.
> But the same problem occurs.
> When using madfuller package, are there any other packages
> also required?
> I use the levinlin package, it works fine.
> But madfuller doesn't.
Nick Cox
> > This is the test example for Kit Baum's
> > -madfuller- from SSC. It works fine for
> > me. I suspect some different problem,
> > such as your Stata not being fully up-to-date.
> > -update all- and try again.
www, a.k.a Terence
> > > While using madfuller to test panel unit root,
> > > I use the example provided by this package.
> > > Those commands I use are following:
> > >
> > > use http://fmwww.bc.edu/ec-p/data/hayashi/sheston91.dta,clear
> > > tsset country year, yearly
> > > madfuller rgdppc if country<11, lags(2(2)8)
> > >
> > > But I got error message like 'variable __000001 not found' or
> > > ' variable
> > > year not found'
> > > Does anyone have same experience like this?
> > > Or the commands I set are wrong?
*
* 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/
