Does usermodel fix residual variances to aid in identification?

[Morgan Morgan]

Hello!

I am using usermodel to estimate a single factor model with six indicators.  

The results generated seem to have fixed (or otherwise estimated) the residual variances for all of the six indicators to 0.5.  Is this part of the operation of the usermodel function which is done to aid in model identification?

Many thanks for any help or thoughts!

All the best, Morgan 

The results are below:

[Elliot Tucker-Drob]

I see that the std_genotype estimates are where you are observing the strange behavior (residual variances all .5s).  I believe that the starting values of the residual variances are half the variances (h2s) of the corresponding variables (i.e. half of 1 for STD_Genotype), which would suggest that the model "converged" on the starting values- i.e. did not converge at all. Some possible reasons that for this that I can think of might be that you altered the convergence criteria in your usermodel syntax, that the h2s are extremely low, the S matrix was excessively smoothed, or rGs are so close to 0 that the model is not identified given the data...

--
You received this message because you are subscribed to the Google Groups "Genomic SEM Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to genomic-sem-us...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/genomic-sem-users/775009bb-de1a-40bd-a599-1c9fe9101107n%40googlegroups.com.