|
|
Hi, I've been trying to work this out with some other online discussion boards and made some progress, but I'm still having difficulties getting parts of the analysis to work out. I have the following format in my data:
> ************************************
> Subj Dis Wo Trt PreF1 PreF2 PreF3 Psch
> 1 1 2 1 1 0 0 1
> 1 1 2 2 0 1 0 0
> 1 1 2 3 0 0 1 1
> 2 2 1 1 1 0 0 0
> 2 2 1 2 0 1 0 1
> 2 2 1 3 0 0 1 0
So The Dis and Wo factors are between subjects (Categorical) and thus constant for
the 3 measures in the same subject. All participants see the same 3
Trt options, which I've Dummy Coded with the PreF variables, and PSCH
is the Binary DV.
What I'm interested in are the main effects of Dis, Wo and Trt, but also the interaction of Dis*Trt. I've been trying to work through this with another blog room, and I've gotten down to the following code with the GEE:
GENLIN psych (REFERENCE=LAST) BY dis wo trt (ORDER=ASCENDING)
/MODEL dis wo trt INTERCEPT=YES
DISTRIBUTION=BINOMIAL LINK=LOGIT
/REPEATED SUBJECT=subj WITHINSUBJECT=trt SORT=YES
CORRTYPE=EXCHANGEABLE ADJUSTCORR=YES
/PRINT CPS DESCRIPTIVES MODELINFO FIT SUMMARY SOLUTION.
where
(1) dis wo and trt are categorical independent variables
(2) subj is the subjects variable
(3) trt is a within subjects independent variable
(4) the var-cov matrix is exchangeable
(5) psych is the dichotomous dependent variable
The problem I have though is that if I add an interaction term with the within-subject factor (e.g. dis*trt), I get the following error message: "The Hessian matrix is singular. Some convergence criteria are not satisfied. The GENLIN procedure continues despite the above warning(s). Subsequent results shown are based on the last iteration. Validity of the model fit is uncertain."
Is there anyway to get around this?
Thanks,
D
|
Locked
