Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: VS: RE: mlogit with constraints


From   "Vartia, Niini" <Niini.Vartia@iue.it>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: VS: RE: mlogit with constraints
Date   Fri, 12 Nov 2004 13:06:47 +0100

Hi!
 
Thank you for your suggestion, Scott. However, it did not solve my problem since the comparison group was already multi==1. So, I still do not know what causes the message: redundant or inconsistent constraints.
 
I tried the following set-up of the constraints and these constraints seem to work in the estimation. However, I am not sure whether this is what I want, i.e. to constrain some of the coefficients related to the explanatory variables to be zero. In this case the constrained coefficient is dropped from the equation.
 
constraint define 1 [3] L2debt

constraint define 2 [2]  debttoas23

constraint list
     1:  [3] L2debt
     2:  [2]  debttoas23
 
I would be grateful for any help with the constraints. I would like to know what is the difference between setting up coinstraints like above and my original constraints:  constraint 1 [3] L2debt=0 and constraint 2 [2]  debttoas23=0?

Thank you in advance.

Best regards,

Laura 

-----Alkuperäinen viesti----- 
Lähettäjä: Scott Merryman [mailto:smerryman@kc.rr.com] 
Lähetetty: to 11.11.2004 22:43 
Vastaanottaja: statalist@hsphsun2.harvard.edu 
Kopio: 
Aihe: st: RE: mlogit with constraints



	I suspect you have defined a constraint for the base category (either
	equation 2 or 3).  You could use the basecategory(#) option to change the
	value of dependent variable that is treated as the base category.
	
	Hope this helps,
	Scott
	
	
	> -----Original Message-----
	> From: owner-statalist@hsphsun2.harvard.edu [mailto:owner-
	> statalist@hsphsun2.harvard.edu] On Behalf Of Vartia, Niini
	> Sent: Thursday, November 11, 2004 7:56 AM
	> To: statalist@hsphsun2.harvard.edu
	> Subject: st: mlogit with constraints
	>
	> Dear Statalisters,
	>
	> I trying to estimate a multinomial model with categorical dependent
	> variable (multi) taking values 1,2 and 3. I would like to constrain some
	> of the coefficients related to the explanatory variables to be zero. After
	> defining the following constraints and estimating mlogit model I get the
	> message below:
	>
	> constraint define 1 [3] L2debt=0
	>
	> constraint define 2 [2]  debttoas23=0
	>
	>  mlogit multi bktvolkasvu avhelieuribor m2togdp invr3 wint3 capint3
	> exint3tot3
	> prcosttot3 glvtot3 rdint1 tplkmyr ika loghenk forown L2profit fixtotot
	> debt
	> toas23 L2debt logcov, cluster(yrtunnew) constraints(1-2)
	>
	> redundant or inconsistent constraints
	> r(412)
	>
	> I must be doing something wrong when specifying the constraints. I Would
	> really be grateful for help in setting up the constraints.
	
	
	
	*
	*   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/
	

<<winmail.dat>>




© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index