NONMEM Users Network Archive

Hosted by Cognigen

Re: OMEGA selection

From: Ethan Wu <ethan.wu75>
Date: Wed, 15 Apr 2009 10:36:43 -0700 (PDT)

Nick, in this case, the model fixed effects parameter estimates do make =
sense since the objective of analysis is for simulation,  should I s=
till ignore failing of $COV ? ________________________________=

om Sent: Wednesday, April 15, 2009 12:17:57 PM Subject: Re: [NMusers] O=
MEGA selection Ethan, Do not pay any attention to whether or not =
the $COV step runs or even if the run is 'SUCCESSFUL' to conclude anything =
about your model. Your opinion is not supported experimentally e.g. see htt=
n and references. NONMEM has no idea if the parameters make sense or n=
ot and will happily converge with models that are overparameterised. You ca=
nnot rely on a failed $COV step or a MINIMIZATION TERMINATED message to con=
clude the model is not a good one. You need to use your brains (NONMEM does=
 not have a brain) and your common sense to decide if your model makes sens=
e or is perhaps overparameterised. Nick Ethan Wu wrote: > > =
Dear all, > >  I am fitting a PD response, and the equation goes lik=
e this: > > total response = baseline+f(placebo response) +f(drug re=
sponse) > > first, I tried full omega block, and model was able to con=
verge, but $COV stop failed. > > To me, this indicates that too many p=
arameters in the model. The structure model is rather simple one, so I thin=
k probably too many Etas. > > I wonder is there a good principle of Et=
a reduction that I could implement here. Any good reference? > > =

of Auckland, 85 Park Rd, Private Bag 92019, Auckland, New Zealand n.holfo=
1-6369 (Apr 6-Jul 17 2009)
Received on Wed Apr 15 2009 - 13:36:43 EDT

The NONMEM Users Network is maintained by ICON plc. Requests to subscribe to the network should be sent to:

Once subscribed, you may contribute to the discussion by emailing: