# st: sureg and reg3 for instrumental variables

 From Valérie Orozco To statalist@hsphsun2.harvard.edu Subject st: sureg and reg3 for instrumental variables Date Mon, 31 Jul 2006 14:33:35 +0200

I have to estimate an AIDS (linear) model with instrumental variables.
I have read the statalist faq and one note "query on sureg" from Kit Baum, and another concerning the "sureg" and "reg3" procedures.

I have 2 questions.

1) I have to write some constraints (adding up, homogeneity, symmetry). If I write all the equations I have and all the constraints I have, I have an error "redundant or inconsistent constraints". I know that I have only n-1 equations to estimate, but the constraints depend on all... So I don't really know how to write the constraints or the equations to estimate. The program will probably be written with something like (but here I have certainly too much constraints).

/*example for 4 goods*/
constraint 1 [dw1]_cons + [dw2]_cons + [dw3]_cons + [dw4]_cons = 1
constraint 2 [dw1]ddeflexp + [dw2]ddeflexp + [dw3]ddeflexp + [dw4]ddeflexp = 0

constraint 3 [dw1]dlnp1 + [dw2]dlnp1 + [dw3]dlnp1 + [dw4]dlnp1= 0
constraint 4 [dw1]dlnp2 + [dw2]dlnp2 + [dw3]dlnp2 + [dw4]dlnp2= 0
constraint 5 [dw1]dlnp3 + [dw2]dlnp3 + [dw3]dlnp3 + [dw4]dlnp3= 0
constraint 6 [dw1]dlnp4 + [dw2]dlnp4 + [dw3]dlnp4 + [dw4]dlnp4= 0

/*homogeneity*/
constraint 7 [dw1]dlnp1 + [dw1]dlnp2 + [dw1]dlnp3 + [dw1]dlnp4= 0
constraint 8 [dw2]dlnp1 + [dw2]dlnp2 + [dw2]dlnp3 + [dw2]dlnp4= 0
constraint 9 [dw3]dlnp1 + [dw3]dlnp2 + [dw3]dlnp3 + [dw3]dlnp4= 0
constraint 10 [dw4]dlnp1 + [dw4]dlnp2 + [dw4]dlnp3 + [dw4]dlnp4= 0

/*symmetry*/
constraint 11 [dw1]dlnp2 = [dw2]dlnp1
constraint 12 [dw1]dlnp3 = [dw3]dlnp1
constraint 13 [dw1]dlnp4 = [dw4]dlnp1
constraint 14 [dw2]dlnp3 = [dw3]dlnp2

sureg (dw1 = ddeflexp dlnp1 dlnp2 dlnp3 dlnp4 period dlnyear) /*
*/(dw2 = ddeflexp dlnp1 dlnp2 dlnp3 dlnp4 period dlnyear)/*
*/ (dw3 = ddeflexp dlnp1 dlnp2 dlnp3 dlnp4 period dlnyear)/*
*/ (dw4 = ddeflexp dlnp1 dlnp2 dlnp3 dlnp4 period dlnyear), cons(1-14)
I have the message "redundant or inconsistent constraints". Does someone have an idea?

2 ) The second question concern instrumental variables because I suspect some variables (for example one which is ddeflexp) to be endogeneous. Is it the right method to use reg3 like this :
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)/*
*/ (ddeflexp = d2deflexp dlogPIB dlogrev), mvreg endog(ddeflexp) exog(d2deflexp dlogPIB dlogrev) /* */cons(1-14)

with d2deflexp, dlogPIB and dlogrev some instruments
?

Thank you very much.

Valérie

--

***********************
Valérie Orozco
INRA ESR Toulouse - FRANCE
ESR INRA - BP52627 - 31326 Castanet Tolosan Cedex orozco@toulouse.inra.fr
05-61-28-50-97
***********************

*
* 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/