RE: Reducing ETAs actually decreased OFV

From: Doug J. Eleveld Date: August 11, 2013 technical Source: mail-archive.com
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 ________________________________
Aug 11, 2013 Xinting Wang Reducing ETAs actually decreased OFV
Aug 11, 2013 Doug J. Eleveld RE: Reducing ETAs actually decreased OFV
Aug 12, 2013 Leonid Gibiansky Re: Reducing ETAs actually decreased OFV
Aug 12, 2013 Bill Denney Re: Reducing ETAs actually decreased OFV
Aug 25, 2013 Xinting Wang Re: Reducing ETAs actually decreased OFV
Aug 25, 2013 Bill Denney Re: Reducing ETAs actually decreased OFV
Aug 25, 2013 Leonid Gibiansky Re: Reducing ETAs actually decreased OFV
Aug 26, 2013 Xinting Wang Re: Reducing ETAs actually decreased OFV
Aug 26, 2013 Bill Denney RE: Reducing ETAs actually decreased OFV
Aug 26, 2013 Leonid Gibiansky Re: Reducing ETAs actually decreased OFV