[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: constraints

From   "Joseph Coveney" <>
To   "Statalist" <>
Subject   Re: st: constraints
Date   Wed, 25 Jun 2008 21:00:36 +0900

Chiara Mussida wrote:

yes, but it is related to the same eq specified w r to 2 different cons.
In a mvprobit I have 6 different equations with 6 different constraints:
mvprobit (private = years logptax loginc) (vote = years logptax
           loginc) (pub12 = years loginc):
how could I add different cons to each eq??


It would be like the following:

use, clear
constraint define 1 [private]loginc = 0.4
constraint define 2 [vote]loginc = 0.6
mvprobit (private = years logptax loginc) ///
(vote = years logptax loginc, nocons), constraints(1 2)

If I copied the -mvprobit- command from your last post correctly, it looks
as if you've got two equations (UE and UN) represented thrice each in your

mvprobit(UE=sex age xfamsize loweduc compulsory diploma ///
 urate experience north centre southislands, constraints(1)) ///
 (UN=sex age married xfamsize loweduc compulsory diploma experience) ///
 (UE=sex age newdiploma urate experience) ///
 (UN=sex age famsize loweduc compulsory diploma experience) ///
 (UE=sex married loweduc compulsory diploma experience) ///
 (UN=sex age married nodiploma experience), cons(1-3)

I'm surprised that you hadn't got an error message like

Equation/parameter /private/ multiply defined

in a pilot run prior to attempting applying constraints.

Joseph Coveney

*   For searches and help try:

© Copyright 1996–2017 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index