Statalist


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

Re: st: likelihood ratio test after mi


From   Maarten buis <maartenbuis@yahoo.co.uk>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: likelihood ratio test after mi
Date   Wed, 9 Dec 2009 10:32:36 +0000 (GMT)

--- On Wed, 9/12/09, raoul reulen <r.c.reulen@gmail.com> wrote:
> mi test and mi testtransform will allow me to test linear
> and non-linear hypotheses within the same model, but what if I
> wanted to compare two nested models? Essentially i have two models,
> one with a linear variable in it and the other with the same variable
> fitted but as an indicator (aka dummy) variable. By comparing the two
> models I can work out whether it is appropriate to fit the variable
> as a linear term. 

If the models are nested, then there exists a contstraint that is
equivalent to the comparison of the two models, for example:

*-------------------- begin example ---------------------
sysuse auto, clear
reg price i.rep78 mpg i.foreign
test 2.rep78 = 3.rep78 / 2 = 4.rep78 / 3 = 5.rep78 / 4
*--------------------- end example ----------------------

Hope this helps,
Maarten

--------------------------
Maarten L. Buis
Institut fuer Soziologie
Universitaet Tuebingen
Wilhelmstrasse 36
72074 Tuebingen
Germany

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


      

*
*   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