UUPharmacometrics / PsN

Perl-Speaks-NONMEM
https://uupharmacometrics.github.io/PsN
GNU General Public License v2.0
65 stars 19 forks source link

PsN SCM maynot be optimizing in some cases #248

Open prewr2 opened 1 year ago

prewr2 commented 1 year ago

Sometimes PsN SCM gives successful minimization message without changing the parameters very much from the initial guess (regardless of initial guess of the parameter, especially added covariate effect). If the same model is run independently it can be minimized to a much lower objective function. Sometimes SCMs OFV with a certain covariate effect is > base model OFV

Is this happening due to PsN SCM using $ETAs in the NONMEM code, and is there a way to avoid it?

rikardn commented 1 year ago

Using $ETAs was a bug that has now been fixed in PsN version 5.3.1. This was intended only for linearized models. This should not affect the initial population parameter estimates though. More details would be needed to be able to understand what is going on.