[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
Re: st: Re: panel unit root tests
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
--- Ricardo Gonçalves Silva <email@example.com>
> 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.
> ----- Original Message -----
> From: "James Shaw" <firstname.lastname@example.org>
> To: <email@example.com>
> 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 +
> > Using appropriate estimators for this model
> > (Arellano-Bond, Anderson-Hsiao), I am getting very
> > large estimates for the autoregressive parameters,
> > and B2. In fact, the sum of the estimates for B1
> > B2 approaches 1.0. I believe that one could
> > test H0: B1 + B2 = 0 vs. H1: B1 + B2 < 0 using a
> > panel data unit root test, such as that described
> > Levin et al. (2002), which is implemented in Stata
> > LEVINLIN. Assuming this is correct, what would
> > appropriate command line be? For example,
> > y, lag(0)" or "levilin y, lag(2)." Alternatively,
> > maybe the problem is easier than I imagine and I
> > simply perform a joint test of the estimates for
> > 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
> > http://smallbusiness.yahoo.com/resources/
> > *
> > * For searches and help try:
> > *
> > * 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/statalist/faq
> * http://www.ats.ucla.edu/stat/stata/
James W. Shaw, Ph.D., Pharm.D., M.P.H.
Tobacco Control Research Branch
Behavioral Research Program
Division of Cancer Control and Population Sciences
National Cancer Institute
Do You Yahoo!?
Tired of spam? Yahoo! Mail has the best spam protection around
* For searches and help try: