Thanks for the prompt response. I was not really sure
what the "lags" option meant when used with levinlin.
My understanding of the unit root test is that it
boils down to a test on the AR parameter(s) in a
dynamic model. The simplest such model would be: y =
By_i(t-1). Since "lags(0)" is a valid option, I was
not sure that "lags(2)" would give me what I wanted.
That is, my presumption was that "lags(0)" would test
the AR parameter for a first-order autoregressive
model.
--
Jim
--- Ricardo Gon�alves Silva <[email protected]>
wrote:
> Hi,
>
> No, you cannot perform the joint test, since, under
> H0, the test statistics
> does not have a standard distribution.
> In fact, to test H0: B1 + B2 = 0 vs. H1: B1 + B2 <
> 0, you must type
> levinlin, lags(2), for the Levin et al. (2002) test.
>
> Rick
> ----- Original Message -----
> From: "James Shaw" <[email protected]>
> To: <[email protected]>
> Sent: Saturday, March 12, 2005 2:10 PM
> Subject: st: panel unit root tests
>
>
> > Everyone,
> >
> > I have a dynamic panel data model that is
> specified as
> > follows:
> >
> > y_it = B0 + B1y_i(t-1) + B2yi(t-2) + ... + A_i +
> u_it
> >
> > Using appropriate estimators for this model
> > (Arellano-Bond, Anderson-Hsiao), I am getting very
> > large estimates for the autoregressive parameters,
> B1
> > and B2. In fact, the sum of the estimates for B1
> and
> > B2 approaches 1.0. I believe that one could
> formally
> > test H0: B1 + B2 = 0 vs. H1: B1 + B2 < 0 using a
> > panel data unit root test, such as that described
> by
> > Levin et al. (2002), which is implemented in Stata
> as
> > LEVINLIN. Assuming this is correct, what would
> the
> > appropriate command line be? For example,
> "levinlin
> > y, lag(0)" or "levilin y, lag(2)." Alternatively,
> > maybe the problem is easier than I imagine and I
> could
> > simply perform a joint test of the estimates for
> B1
> > and B2 to see if they are equal to 1.0.
> >
> > --
> > Jim
> >
> >
> >
> >
> >
> > then how would I would
> >
> > James W. Shaw, Ph.D., Pharm.D., M.P.H.
> > Post-Doctoral Fellow
> > Tobacco Control Research Branch
> > Behavioral Research Program
> > Division of Cancer Control and Population Sciences
> > National Cancer Institute
> > Tel.: 215-852-3045
> > Fax: 215-542-7150
> >
> >
> >
> > __________________________________
> > Do you Yahoo!?
> > Yahoo! Small Business - Try our new resources
> site!
> > http://smallbusiness.yahoo.com/resources/
> > *
> > * 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/
> >
> >
> >
> > --
> > No virus found in this incoming message.
> > Checked by AVG Anti-Virus.
> > Version: 7.0.308 / Virus Database: 266.7.1 -
> Release Date: 9/3/2005
> >
> >
>
>
>
> --
> No virus found in this outgoing message.
> Checked by AVG Anti-Virus.
> Version: 7.0.308 / Virus Database: 266.7.1 - Release
> Date: 9/3/2005
>
> *
> * 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/
>
James W. Shaw, Ph.D., Pharm.D., M.P.H.
Post-Doctoral Fellow
Tobacco Control Research Branch
Behavioral Research Program
Division of Cancer Control and Population Sciences
National Cancer Institute
Tel.: 215-852-3045
Fax: 215-542-7150
__________________________________________________
Do You Yahoo!?
Tired of spam? Yahoo! Mail has the best spam protection around
http://mail.yahoo.com
*
* 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/
