Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
st: GLLAMM for multinomia logisic
From
"Pesola, Francesca" <[email protected]>
To
"[email protected]" <[email protected]>
Subject
st: GLLAMM for multinomia logisic
Date
Tue, 25 Mar 2014 16:19:36 +0000
Hi STATAlisters,
I have used gllamm to run a multinomial logistic regression with a 3-category DV and a 3-category predictor while accounting for clustering at the individual level (for repeated measures). I used the following syntax:
xi: gllamm sat3gw i.topic3g_w, i(ID) init base(1) link(mlogit) family(binomial) which appears to run correctly:
Condition Number = 5.4910233
gllamm model
log likelihood = -2361.9757
------------------------------------------------------------------------------
sat3gw | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
c2 |
_Itopic3g_~2 | -.0239435 .1367381 -0.18 0.861 -.2919452 .2440583
_Itopic3g_~3 | .0903968 .1999332 0.45 0.651 -.3014651 .4822587
_cons | -.9658663 .0675027 -14.31 0.000 -1.098169 -.8335635
-------------+----------------------------------------------------------------
c3 |
_Itopic3g_~2 | -.0107605 .1162895 -0.09 0.926 -.2386837 .2171626
_Itopic3g_~3 | .1816609 .167406 1.09 0.278 -.1464489 .5097707
_cons | -.4975142 .0576402 -8.63 0.000 -.610487 -.3845414
------------------------------------------------------------------------------
In order to an overall test to assess whether the predictor is associated with the outcome, I ran
test _Itopic3g_w_2 _Itopic3g_w_3
The output suggests that the chi2 associated with the test command has 4 degress of freedom, is it because it compares _Itopic3g_w_2 _Itopic3g_w_3 for levels c2 and c3 (vs. c1)?
. test _Itopic3g_w_2 _Itopic3g_w_3
( 1) [c2]_Itopic3g_w_2 = 0
( 2) [c3]_Itopic3g_w_2 = 0
( 3) [c2]_Itopic3g_w_3 = 0
( 4) [c3]_Itopic3g_w_3 = 0
chi2( 4) = 12.27
Prob > chi2 = 0.0155
Grateful for your help and any reference for gllamm with multinomial logistic :)
Thanks for your help,
Francesca
*
* 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/