Sebastien,
Suppose you define your constraints ahead of the equation.
constraint 1 [dep var]inf.=0
constraint 2 [dep var]var2test=0
then your var command should have in it a varconstraint(1/2) option.
Regards,
Bob
Robert A. Yaffee, Ph.D.
Research Professor
Silver School of Social Work
New York University
Biosketch: http://homepages.nyu.edu/~ray1/Biosketch2008.pdf
CV: http://homepages.nyu.edu/~ray1/vita.pdf
----- Original Message -----
From: Sebastian Kruk <[email protected]>
Date: Saturday, December 27, 2008 9:39 am
Subject: Re: st: restricted VAR for preselection of instruments
To: [email protected]
> Bob,
>
> I have to estimate a VAR with two equations:
>
> (eq1) f.inf = a1 + a10*inf + a11*l.inf + a12*l2.inf + ... + a1j*lj.inf
> + b10*x + b11*l.x + b12*l2.x + ... + b1i*li.x
> (eq2) st = a2 + a21*l.st + a22*l2.st + ... + a2j*lj.st + b20*x
> +b21*l.x + b22*l2.x + ... + b2i*li.x
>
> where x can be hp_gap, cf_gap or bk_gap.
>
> I try a new one:
>
> foreach x of varlist hp_gap cf_gap bk_gap {
> forvalues j=10(-1)1 {
> var f.inf st, exog(l2.`x' l.`x' `x') lags(2/`j') lutstats
> est store mod`x'2_`j'
> }
> }
> est stats _all
>
> But I need next restriction: a10=0
>
> What can I do?
>
> Bye,
>
> Sebastian.
> *
> * For searches and help try:
> * http://www.stata.com/help.cgi?search
> * http://www.stata.com/support/statalist/faq
> * http://www.ats.ucla.edu/stat/stata/
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
