Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down at the end of May, and its replacement, statalist.org is already up and running.


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

Re: st: ST: constraining parameters between models


From   Maarten Buis <maartenlbuis@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: ST: constraining parameters between models
Date   Mon, 12 Nov 2012 16:22:21 +0100

On Mon, Nov 12, 2012 at 3:59 PM, Ronald McDowell wrote:
> Is it possible to constrain parameters to be equal between models in Stata? For example, I have an ordered logistic regression I wish to run for 3 groups and would like to constrain the coefficient of a predictor to be the same in each group on theoretical grounds.
>
> When doing this with linear or logistic regression, I am able to use dummy variables and use the , noconst option to get the desired results simultaneously for each group.

There is no fundamental difference between a model with 3 indicator
variable (and thus no reference category) and no constant and a model
with 2 indicator variables (and thus a reference category) and a
constant. See: M.L. Buis (2012) "Stata tip 106: With or without
reference", The Stata Journal, 12(1), pp. 162-164.

So if you can get this with appropriate indicator variables in -logit-
you can also do it in -ologit-, the only thing you need to do is leave
out the reference category.

-- Maarten

---------------------------------
Maarten L. Buis
WZB
Reichpietschufer 50
10785 Berlin
Germany

http://www.maartenbuis.nl
---------------------------------

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/


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