Output for multinomial logisitc regre... PreviousNext
Mplus Discussion > Multilevel Data/Complex Sample >
Message/Author
 Brian Hall posted on Tuesday, February 23, 2010 - 10:10 am
Dear list:
I am running a multinomial logistic regression utilizing NORM and combining 10 data sets for multiple imputation.

I would like to have the following output from Mplus, if possible:
1. -2log likelihood for each variable
2. chi square test for variable inclusion
3. and the mean for each variable for each level of the DV, including the reference category.

I am calculating the EXP(B) in excel, but I would also like to compute the 95% CI about the OR. Any advice about how to calculate this?

Thanks very much in advance for the help.
 Bengt O. Muthen posted on Tuesday, February 23, 2010 - 4:15 pm
1. If you mean for each imputed data set, no Mplus does not give this automatically.

2. If you mean the significance of a certain covariate, no, Mplus does not give this automatically; you will have to square the z value.

3. Mplus does not give this.

You can do exp(b) in model constraint and then get the CI from those SEs. Or, the more common way, get the SEs and the CIs from the b's and then exponentiate the CI limits.
Back to top
Add Your Message Here
Post:
Username: Posting Information:
This is a private posting area. Only registered users and moderators may post messages here.
Password:
Options: Enable HTML code in message
Automatically activate URLs in message
Action: