Dear NM Users, I have a question regarding EVID=2 and the covariance step. I am fitting a tumor growth inhibition model to my data and I am still at the beginning where I am using a base model (no covariates). I have time changing covariates in the dataset which I am working with, some of which were measured at time points which do not coincide with the DV observations and hence I used EVID=2.
When I first run the model (FOCE+I), I got very high relative standard errors for the parameter estimates. However when I duplicated the control stream with just one change : IGNORE=(EVID.EQ.2), I got exactly the same OFV and the same parameter estimates (this is obvious since I have not included any of the covariates), but with dramatically lower relative standard errors. Do you have an explanation for what may have went wrong ? (by the way, I used FO just for testing and in both cases the RSEs were exactly the same). Best Regards, Ahmed Abbas Suleiman, MSc PhD Candidate at Bonn University Institute of Pharmacology - Clinical Pharmacology Unit Hospital of the University of Cologne Gleueler Str. 24, 50931 Cologne, Germany Phone: +49-(0)-221-478-3402
