Statalist


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

Re: st: Specify coefficient vectors


From   "Austin Nichols" <austinnichols@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: Specify coefficient vectors
Date   Tue, 3 Jul 2007 14:18:27 -0400

I don't think what you were doing before was what you think it was.
The -from- option passes a starting point to the ML process, and
-iterate(0)- says stop after the first iteration. But nothing in that
gives you a real e(V) matrix nor allows you to do any post-estimation
testing.
Perhaps you want constrained regression, but if you specify enough
constraints, the resulting e(V) will have a bunch of zeros, also not
allowing you to do any post-estimation testing.

sysuse auto, clear
constraint 2 price = 1
constraint 2 weight = 2
cnsreg mpg price weight, constraints(1 2)
mat li e(V)

On 7/3/07, K.Boughey@uea.ac.uk <K.Boughey@uea.ac.uk> wrote:
Dear statlist,

How do I specify coefficient vector values using -regress-? I would like
to test a linear regression model estimated using one dataset on a new
dataset, and run various post estimation tests. I have achieved this in
the past when using -clogit- with -iterate-, ie

clogit depvar indepvars, group() iterate(0) from(indepvar=#)

but iterate does not work with -regress-. Is there an alternative?

*
*   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/



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