Message/Author 

Yohan Choi posted on Sunday, September 02, 2018  2:41 pm



how to implement twolevel multinomial logistic model with separate but correlated random effects in Mplus? ex) y = (1, 2, 3) independent variabels = x1, x2 cluster = pid In stata, gsem (2.y < x1 x2 M1[pid]) /// (3.y < x1 x2 M2[pid]), /// cov(M1[pid]*M2[pid]) mlogit (in sem manual in Stata, example 41g) How to implement this in Mplus? 


Try using these pieces in a Type=Twolevel run (see UG chapter 9 for general twolevel setups) using the example of a 3category nominal y variable. Variable: Nominal = y; within = x1 x2; Model: %within% y on x1 x2; %between% y#1 with y#2; The between level can also regress the random effects y#1 and y#2 on a betweenlevel covariate. or be used to predict some betweenlevel DV (don't think Stata can do that). 

Yohan Choi posted on Monday, September 03, 2018  7:21 pm



Thanks professor. Additionally, I wonder that base category of y must be largest number. Is it fixed in Mplus or changeable? 


Use Define to rescore the variable and thereby change the base category. 

Yohan Choi posted on Thursday, September 13, 2018  9:38 pm



Hello, Professor. I did like the following, but confronted problems. I represent a part. Y is (1,2,3,4,5) Variable: usevariables=pid y yg11yg14 yic11yic14 x12x31; cluster=pid; nominal=y; within=yg11yg14 yic11yic14 x12x31; Analysis: Type=twolevel; Model: %within% y on yg11yg14 yic11yic14 x12x31; %between% y#1 with y#2 y#3 y#4; y#2 with y#3 y#4; y#3 with y#4; *** WARNING in MODEL command One or more betweenlevel covariances involving the following variable is free while its betweenlevel variance is fixed at 0. Fix all betweenlevel covariances with this variable at 0 or free its variance. Problem with: Y#1 it's same for Y#2, Y#3 Between Level Y#1 WITH Y#2 0.000 Y#3 0.000 Y#4 0.000 Y#2 WITH Y#3 0.000 Y#4 0.000 Y#3 WITH Y#4 0.000 Means Y#1 4.226 Y#2 8.726 Y#3 8.248 Y#4 3.603 And I need constant's coefficients, but did not calculate that. Sorry for seemingly basic questions. I really need your help, professor. 


Add the mentioning of their variances: y#1  y#4; But note that this results in 4dimensional integration which is heavy, slow, and imprecise (you can try Integration = Montecarlo(5000). You may want to put a factor behind these 4 as shown in UG examples for 2level mixture modeling (the latent class variable is also nominal). 

Yohan Choi posted on Friday, September 14, 2018  6:57 pm



Thanks professor. Very thanks for your kind reply. I have one more question. Can I have predicted probability with 95% CI with some fixed covariates in this model? 


You can get that using Model Constraint to express the estimated logits in terms of probabilities (see e.g. end of UG chapter 14). 

Yohan Choi posted on Monday, September 17, 2018  6:12 pm



Thanks Muthen. There are another proper methods to reduce convergence time? 


Only the 2 suggestions I made. 

Back to top 