Attachment | Size |
---|---|
post.txt [6] | 8.62 KB |
Hello! This is a follow-up post to some guidance I got on this board awhile ago that was incredibly helpful. I am trying to run a bivariate saturated model that controls for age and sex. The two variables were assessed at different time points, so one variable needs to be adjusted for age1 and sex, the other variable for age2 and sex.
I haven't found a script that does this, so I attempted to adapt an existing one. The entire purpose of running the saturated model is to 1) get cross-twin cross-trait correlations and 2) check assumptions to be sure everything can be constrained. That's all I want to do! My script for running the bivariate Cholesky seems to work fine.
I assume that this analysis would be fairly common given that assumption checking is part of the process and CTCT correlations are often reported. But, I'm just having a ton of problems with actually running it. Lots of Mx RED error codes, and then when I get to the final steps (constraining means, variances, and covariances, and then constraining CTCT), I am getting error code 10 about implausible start values. Sometimes when I use mxTryHard this goes away, other times it does not and there are errors, and still other times it takes so long that the system crashes.
Granted I'm still a novice, but I have never ever had so many errors (or error code 10), so I'm just wondering if I'm missing something obvious that I've done wrong or if there's a way to make my code less computationally intensive, as I have to repeat this analysis many times.
Any help would be much appreciated.