Dear Stata List,
I have written a program to compute standard errors (corrected for
choice-based sampling) for a conditional logit model. The first few lines
of my program are:
capture program drop wesmle
program define wesmle, eclass
xi: clogit `1'
local n `e(N)'
local ll0 `e(ll_0)'
local logl `e(ll)'
local freepar `e(df_m)'
local choice `e(depvar)'
local idv `e(group)'
tempname coef covar
matrix `coef'=e(b)
matrix `covar'=e(V)
-- and so forth.
However, I need to constrain the value of one of my coefficients to be
fixed (at, say, -6). The "eclass" will not allow for constrained
estimation. Is there any way of getting around this? Can I access the
system variables (e.g., e(V)) without defining my program as an eclass?
Many thanks!
Elizabeth
*
* 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/
