Hello,

I am currently trying to run a joint longitudinal survival model using the stjm command. I am looking at modeling a longitudinal biomarker (in the code below "log10_cy") with time to a cancer diagnosis. This biomarker is measured as a cumulative exposure, so the slope for each individual can only increase over time. The data have been correctly stset. I would ideally like to allow for shared intercept and slopes, and am using the rfp option accordingly.

When I run the model using the code below, I receive this error: " BLUP calculation failed Variance of BLUPs for a specific panel could not be estimated Try increasing the number of quadrature nodes" r(1986). I have tried increasing the quadrature nodes using the gh() option, and also clicked on "r(1986)" but there are no entries. Log has also been attached for reference.

stjm log10_cy , panel(naid) rfp(1) covariance(un) ///
survm(weibull) survcov(raceeth) ///
intassociation nocurrent nocoefficient gh(30)

Has anyone else had this type of error when running stjm? Alternatively, does anyone has any suggestions for troubleshooting this error to see if I can find the reason why it is not running (e.g. a way to find out which panel/s are causing the problem)? Any advice would be appreciated!