Andrea
You may want to look at my Stata Journal paper, where the
assumptions required by -xtlsdvc- are spelled out in detail:
http://ideas.repec.org/a/tsj/stataj/v5y2005i4p473-500.html
or its unpublished version
http://ideas.repec.org/p/cri/cespri/wp165.html
Just a quick answer here: strictly exogenity is a required
assumption for the current version of -xtlsdvc-. Failing that,
the bias approximation computed by the code is not correct.
Giovanni
Scrive sistoand80 <[email protected]>:
> Hi, could any one explain me what happens when, using xtlsdvc and
> initializing with a consistent estimator (say, for example, Arellano and
> Bond) the diagnostic tests fail to support for fully exogeneity of the
> instrument set or absence of AR(2) serial correlation? (more precisely,
> Sargan test's p-value is 0.000 and AB AR(2) test fail to reject the presence
> of AR2 serial correlation, suggesting several problem in my instrument set).
> Are the corrected lsdv estimates biased? Are consistent...My problem is that
> I suppose that a partition of my instrument set is to be treated as
> predetermined (I've checked with xtabond, setting some variables as
> predetermined and verifying the robustness of the diagnostic tests) but
> xtlsdvc treates all regressors as strictly exogenous. How can I deal with
> this problem?
> Thank you in advance for the replay
>
> Andrea Sisto
> University of Turin (Italy)
>
>
>
> ------------------------------------------------------
> Passa a Infostrada. ADSL e Telefono senza limiti e senza canone Telecom
> http://click.libero.it/infostrada20dic06
>
>
>
> *
> * 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/
>
--
Giovanni S.F. Bruno
http://ideas.repec.org/e/pbr136.html
Istituto di Economia Politica, Universit� Bocconi
Via U. Gobbi, 5, 20136 Milano
Italy
tel. + 02 5836 5411
fax. + 02 5836 5438
*
* 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/
