search

Genentech / jmpost

https://genentech.github.io/jmpost/
17 stars 4 forks source link

Double check initial values for SF and GSF models #349

Closed gowerc closed 5 months ago

gowerc commented 5 months ago

Whilst developing the Claret-Bruno model I noticed that we had a few cases where exp(values) were being passed to places where they should have been logged. Just double check to see if these models are having the same issue. Particularly with regards to the population means

gowerc commented 5 months ago

One thing that definitely needs fixing is that simulation functions have all of the mu parameters on the exp or observation scale whereas they should be on the log scale.

On reflection I think the overall implementation in GSF and SF is correct, just need to double check the initial values