

Covariates with zero variance in LCGA 

Message/Author 


Variable: names are id bsidep1 bsidep2 bsidep3 bsidep4 x1 x2 x3; usevar = bsidep1bsidep4 x1 x2; missing = all (999); CLASSES = c(4); SAVEDATA: FILE IS dep_4q_x1x2_output; save=cprobabilities; Analysis: type = MIXTURE missing; STARTS = 1000 300; STITERATIONS = 20; Model: %OVERALL% i s q bsidep1@0 bsidep2@1.5 bsidep3@5.5 bsidep4@11.5; iq@0; c#1 ON x1 x2; c#2 ON x1 x2; c#3 ON x1 x2; Output: sampstat standardized tech1 tech11 tech14; This model runs fine, but I get this error message: WARNING: THE SAMPLE VARIANCE OF X1 IN CLASS 2 IS 0.000. (the same for X2sf) ONE OR MORE MULTINOMIAL LOGIT PARAMETERS WERE FIXED TO AVOID SINGULARITY OF THE INFORMATION MATRIX. THE SINGULARITY IS MOST LIKELY BECAUSE THE MODEL IS NOT IDENTIFIED, OR BECAUSE OF EMPTY CELLS IN THE JOINT DISTRIBUTION OF THE CATEGORICAL LATENT VARIABLES AND ANY INDEPENDENT VARIABLES. THE FOLLOWING PARAMETERS WERE FIXED: 23 22 This was the gamma parameters for class 2 on to x1 and x2. Does fixing the regression coefficient of the latent construct (class) on to the exogenous variable (x1 and x2) make this model uninterpretable? 


This is not a problem. It just tells you that all members of class 2 have the same value for x1. 


Great, thank you very much! Sheila 

Back to top 

