iam a novice modeler, iam getting this error constantly, irrespective of the error parameters and compartments, iam not caculate the variance covariance matrix while trying to run
infer(fit) , ive also tried
infer(fit1,sandwich_estimator=false)
The error message says
Variance-covariance matrix could not be evaluated.
The random effects may be over-parameterized.
Check the coefficients for variance estimates near zero.
And checking your parameter estimates table, we see that 3 out of you 4 Omegas are < 0.033.
This gives Etas that are around 3%:
julia> exp(0.033)
1.0335505392413056
So, you might want to follow through with the error message suggestion and revisit the Omegas that might be over-parameterized?
Additionally, if you are confident that these Omegas don’t need change, then you can try a different estimation method for the confidence intervals.
A good approach is to use a non-parametric one (that does not make assumptions about the underlying distribution of you parameters) such as Bootstrap().
Try:
infer(fit, Bootstrap())