RE: Reducing ETAs actually decreased OFV
You mean you removed an eta and the objective function went down? I dont think
this can really happen in a straightforward way.
In NONMEM, the minimum of the objective function is found and if having all
etas to zero gives a lower objective function than some other eta values then
barring covergence problems all zero etas should have been found.
warm regards,
Douglas Eleveld
Quoted reply history
________________________________________
From: [email protected] [[email protected]] on behalf of
Xinting Wang [[email protected]]
Sent: Sunday, August 11, 2013 4:23 AM
To: [email protected]
Subject: [NMusers] Reducing ETAs actually decreased OFV
Dear all,
Does anyone witnessed such a phenomenon in NONMEM as when you reduced an ETA,
the OFV value, rather than increase, actually decreased? It's quite against
intuition, as individual estimation should be better than population estimation
in that particular parameter. Both models, whether having this ETA, converged
very well.
Best
--
Xinting
________________________________