Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at

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

st: Re: Margins and interactions

From   Arti Rayit <>
Subject   st: Re: Margins and interactions
Date   Fri, 13 May 2011 12:15:02 +0100 (BST)

I have a longitudinal experiment with 3 factors, a (3 levels) b (2 levels) and c 
(2 levels) . As there is missing data and the data is not collected at regular 
time intervals I have used xtmixed to analyse the data:

xtmixed sc II pid:, ml
margins a##b##c, post  asbalanced emptycells(reweight) 

I would like to test if the a#b and possibly other interactions are significant. 
Is it acceptable to fit the model with and without the a#b interaction and 
compare the results with lrtest? I ask this because I understand that the way 
the factors are coded can cause some difficulties in the interpretation of 

Thank you,


*   For searches and help try:

© Copyright 1996–2017 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index