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 on April 23, and its replacement, statalist.org is already up and running.


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

Re: st: Interaction in Generalized Ordered Logistic regression_gologit2 (ORs and CIs)


From   Richard Williams <richardwilliams.ndu@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: Interaction in Generalized Ordered Logistic regression_gologit2 (ORs and CIs)
Date   Wed, 18 May 2011 10:19:49 -0500

If I understand the question, I believe you would do this the same way as you would if you were using mlogit. Something like

use "http://www.indiana.edu/~jslsoc/stata/spex_data/ordwarm2.dta";, clear
gen yr89ed = yr89 * ed
gologit2  warm yr89 male white age ed prst  yr89ed, npl(yr89ed) lrf
forvalues i = 1/3 {
lincom [#`i']ed + [#`i']yr89ed, or
}

The key thing is you have to specify the equation #s for lincom; otherwise it only does the first one.


At 04:29 AM 5/18/2011, Sanam P wrote:
Dear Stata list

I have done a partial proportional odds regression (Generalized Ordered Logistic
Regression) using gologit2 command with most of categorical variables having
more than 2 categories:

xi: gologit2   y  i.a   i.b  i.c, autofit(0.01) lrforce or

I would like to include interaction terms between one of the covariates which
has 2 categories with other covariates in the model.

Like other regression models I have tested for significance of the interactions
using Likelihood Ratio tests:

e.g.:
xi: gologit2   y  i.a *i.b  i.c, autofit(0.01) lrforce or
 eststore A
xi: gologit2   y  i.a i.b  i.c, autofit(0.01) lrforce or
 eststore B
 lrtestA B

 For the interactions that are significant, I would like to have the ORs and
Confidence intervals for the effect of variables included in the interaction in
each panel.

In other regression models it is possible to calculate for example Hazards
Ratios and confidence intervals for the interactions, using Lincom command.

for example for the effect of age in stages 2 and 3 of cancer (Hrs and CIs),
after running the model using stcox:
**effect of age for stage 2
 lincom age+ _IstaXage_d_2,hr
**effect of age for stage 3
 lincom age+ _IstaXage_d_3,hr

However in Generalized Ordered model there are more than one OR for each
covariate (depending on the number of levels of the outcome variable which will
determine number of the panels), and lincom command will give only one OR and
one CI for each variable's effect..

I would really appreciate it if you could help me on how I can have the ORs and
CIs for the effect of the variables included in the interaction terms for all
the panels in the generalized Ordered Logistic model in stata 11.

*
*   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/

-------------------------------------------
Richard Williams, Notre Dame Dept of Sociology
OFFICE: (574)631-6668, (574)631-6463
HOME:   (574)289-5227
EMAIL:  Richard.A.Williams.5@ND.Edu
WWW:    http://www.nd.edu/~rwilliam

*
*   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   |   Site index