Cross Classified Multinomial Logit PreviousNext
Mplus Discussion > Multilevel Data/Complex Sample >
 Chris Weber posted on Sunday, August 26, 2012 - 8:06 am

I am attempting to fit a multinomial logit (3 categories), with random effects. The intercepts vary across news organizations and congressional races. However, news organizations are not perfectly nested within races, leading a cross classified structure. My question is whether this type of model can be fit in v6.12. If so, is the model I include below accomplishing this goal, or is it assuming observations are nested within news organization nested within races? I have not been able to find documentation on cross classified models in mplus


Margin days_elec race_id Org_Num
offyear corporate female_candidate
gov_race open_seat logcirc incumbent_years
char femcanXgov;
WITHIN= corporate days_elec logcirc
Margin offyear female_candidate
gov_race open_seat incumbent_years femcanXgov ;
CLUSTER are race_id Org_Num;
nominal is char;

missing are all (999);

type=complex twolevel;

char#1 char#2 on Margin days_elec corporate logcirc offyear female_candidate
gov_race open_seat incumbent_years femcanXgov;

char#1 char#2 ;
 Linda K. Muthen posted on Sunday, August 26, 2012 - 8:08 am
Cross-classified models are not available in Version 6.12. They will be available in Version 7.
Back to top
Add Your Message Here
Username: Posting Information:
This is a private posting area. Only registered users and moderators may post messages here.
Options: Enable HTML code in message
Automatically activate URLs in message