Message/Author 

jan mod posted on Tuesday, August 25, 2015  6:57 am



Dear, I have a question. Is it possible to do multinomial multilevel (random intercept) with my categorical dependent variable with three categories? I cannot find syntax examples for this type of analysis. I use the nominal statement. 


Yes, that is possible. On level 2 you have C1 continuous latent variables (random intercepts) for a nominal with C categories. For a variable names y they are referred to as y#1, y#2, y#3. 

jan mod posted on Tuesday, August 25, 2015  11:00 am



Thanks! Do you also have a reference that I can use to write the equations. Multilevel multinomial is new for me. 


I am running a twolevel binary logistic model with "INTEGRATION = MONTECARLO; ALGORITHM = INTEGRATION". I get this error message: ONE OR MORE PARAMETERS WERE FIXED TO AVOID SINGULARITY OF THE INFORMATION MATRIX. I am not able to increase the start values becuase it may take a week or two with my current computer to complete that analysis (which may need to be repeated a dozen of times). Can you suggest an appropriate algorithm to accelerate the analysis, or a way to handle this error message without increasing analysis time, any quick fixes? many thanks in advance 


We need to see your full output; send it including Tech1 and Tech8 to Support along with your license number. 

Back to top 