st: Re: overidentification test after reg3

[Kit Baum <baum@bc.edu>]

Valerie,

I don't think that the reg3 syntax you are using will properly handle  
the notion that a nonlinear transformation of a LHS variable should  
be considered endogenous. The alternative syntax -- with the endog()  
option -- will allow you to specify that those d ln p variables are  
to be treated as endogenous. But the default syntax of reg3 (see help  
reg3) causes Stata to take the LHS variables as endogenous and treat  
all others as exogenous. You will probably get quite different  
results from your reg3 if you correct this. I would suggest  
estimating the system without constraints, as you have it and as it  
should be, to analyse the differences, before reimposing the  
constraints appropriate from economic theory.

Others may want to comment on the notion of an instrumental variables  
estimator in which some of the endogenous variables to be  
instrumented are nonlinear transformations of the LHS variables.

The Wooldridge formula you quoted for the Sargan test should then  
give you a somewhat more sensible d.f. if all the references to d ln  
p variables are taken as endogneous variables rather than exogenous  
variables.

Kit Baum, Boston College Economics
http://ideas.repec.org/e/pba1.html
An Introduction to Modern Econometrics Using Stata:
http://www.stata-press.com/books/imeus.html


On Oct 1, 2006, at 2:33 AM, statalist-digest wrote:


> Thank you for your trying to help me. What I didn't mentionned (and  
> that's =
> why
> it was unclear) is that the
>
> dlnp`i' with i =3D1, 2, 3, 4 which are on the RHS of the first 4  
> equations
> correspond to the p`i' (dlnp`i' are just in log and in difference).
> ddeflexp is also a tranformed form for exp.
> So the variables I supposed to be endogeneous are the four dlnp`i'  
> and exp,
> that's why I have written 5 equations of instrumentation.
>
> val=E9rie
>
>
> -  
> ---------------------------------------------------------------------- 
> -----=
> - -----
>
> > From   Kit Baum <baum@bc.edu>
> To   statalist@hsphsun2.harvard.edu
> Subject   st: Re: overidentification test after reg3
> Date   Thu, 28 Sep 2006 11:00:47 -0400
>
> -  
> ---------------------------------------------------------------------- 
> -----=
> - -----
> It is not apparent that this system is simultaneous. The dependent  
> variable=
> s are
> dw1,dw3, dw3, exp, p1, p2, p3, p4 =3D 7. I do not see any of those  
> variable=
> s on
> the RHS of other equations. If that is the case, there is no  
> simultaneity h=
> ere,
> and each equation can be estimated by OLS (or, with constraints across
> equations, as a SUR via sureg).
>
> If there is no simultaneity, there is no question of  
> identification, and th=
> us no
> need for a test of overidentifying restrictions. It is not  
> surprising that =
> the
> routine referenced should come up with a huge number of d.f., as  
> the number=
>   of
> exogenous variables is huge, and that will be multiplied by 7. By  
> my count
> you're estimating 48 RHS coefficients (plus constant terms) minus 6
> constraints, or 42. Somehow, I knew the answer should be 42.
>
>
> Kit Baum, Boston College Economics
> http://ideas.repec.org/e/pba1.html
> An Introduction to Modern Econometrics Using Stata:
> http://www.stata-press.com/books/imeus.html
>
>
> On Sep 28, 2006, at 2:33 AM, Valerie wrote:
>
>
>
> am trying to run an over identification test for a constrained
> simultaneaous model (with reg3). It is the following :
>
> /*symmetry*/
> constraint 1 [dw1]dlnp2 =3D [dw2]dlnp1
> constraint 2 [dw1]dlnp3 =3D [dw3]dlnp1
> constraint 3 [dw2]dlnp3 =3D [dw3]dlnp2
> /*homogeneity*/
> constraint 4 [dw1]dlnp1 + [dw1]dlnp2 + [dw1]dlnp3 + [dw1]dlnp4 =3D0
> constraint 5 [dw2]dlnp1 + [dw2]dlnp2 + [dw2]dlnp3 + [dw2]dlnp4 =3D0
> constraint 6 [dw3]dlnp1 + [dw3]dlnp2 + [dw3]dlnp3 + [dw3]dlnp4 =3D0
>
> reg3(dw1 ddeflexp dlnp1 dlnp2 dlnp3 dlnp4 period dlnyear) /*
> */ (dw2 ddeflexp dlnp1 dlnp2 dlnp3 dlnp4 period dlnyear)/*
> */ (dw3 ddeflexp dlnp1 dlnp2 dlnp3 dlnp4 period dlnyear) /*
> */ (exp explag PIB period)/*
> */ (p1 pxlaitcomteinterp pxlaitcomteinterplag1 ind_sal_interp
> pourc_vol_ean_1 pourc_vol_MDDHD_1)/* */ (p2 pxlaitstandinterp
> pxlaitstandinterplag1 ind_sal_interp pourc_vol_ean_2  
> pourc_vol_MDDHD_2)/*
> */ (p3 pxlaitcomteinterp pxlaitcomteinterplag1 pxlaitstandinterp
> pxlaitstandinterplag1 ind_sal_interp pourc_vol_ean_3 pourc_vol_MDDHD_3
> pourc_vol_parm pourc_vol_beauf) /*
> */ (p4 pxlaitstandinterp pxlaitstandinterplag1 ind_sal_interp
> pourc_vol_ean_4 pourc_vol_MDDHD_4), cons(1-6)
*
*   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/