Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: Linear expenditure system with NLSUR


From   PEREZ PEREZ JORGE EDUARDO <perez.jorge@ur.edu.co>
To   "statalist@hsphsun2.harvard.edu" <statalist@hsphsun2.harvard.edu>
Subject   st: Linear expenditure system with NLSUR
Date   Tue, 30 Jun 2009 22:35:18 +0000

Hi,

I'm trying to estímate a 7-equation linear expenditure system by ML using the nlsur, ifgnls command. Here's the code I'm using

glo NEQN = 7   /* Number of equations */
glo NM=$NEQN-1
global subs "price1*{gamma1=100000}"
forvalues i=2(1)$NEQN {
global subs "$subs + price`i'*{gamma`i'=100000}"
}
global com ""
forvalues i=1(1)$NM {
global com "$com (expenditure`i' = price`i'*{gamma`i'=100000} + {beta`i'} * (totalexpenditure - ($subs)) )"
}
nlsur $com, vce(robust) ifgnls

 The command should estimate 6 "beta" parameters and 7 "gamma" parameters.  However, even though I've omitted one of the equations from the system, STATA does not estimate the last gamma parameter, returning the initial value and a missing standard error.


How can I handle this? Am I doing anything wrong? I also tried to estimate it by ml using the ml commands, but I tried to use method d0, and it doesn't allow me to provide initial values for the gamma parameters.

Thanks in advance,

Jorge Eduardo Pérez Pérez

*
*   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/



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