Re: [NMusers] how to implement AR residual error model in Nonmem with FOCE INT routine

Dear Marc,   This is really helpful!   The model is up running now=
.  One thing I noticed is that FOCE INT with AR model takes really long=
 time, so far it is still running after 4 hr. I also tried FOCE with AR, it=
 coverged very quickly.  About your comment on utility  of AR mode=
l,    in my case, I really can't find a better structure model, and f=
or the later covariate modeling, as I understand from a paper from Mats' =
group,  having a proper residual model could prevent inflation of t=
ype1 error.     ________________________________ Fr=
om: "Gastonguay, Marc" <marcg_at_metruminstitute.org> To: Ethan Wu <ethan.wu=
75_at_yahoo.com> Cc: nmusers_at_globomaxnm.com Sent: Tue, December 29, 2009 9=
:44:28 AM Subject: Re: [NMusers] how to implement AR residual error model=
 in Nonmem with FOCE INT routine Hello Ethan, This archived NMusers=
 discussion thread might be useful: http://cognigencorp.com/nonmem/nm/=
97dec172003.html Marc Marc R. Gastonguay, Ph.D. President & CEO=
, Metrum Research Group LLC  < metrumrg.com > Scientific Director, Metr=
um Institute < metruminstitute.org > 2 Tunxis Rd, Suite 112, Tariffville,=
 CT 06081  Direct: +1.860.670.0744  Main: +1.860.735.7043  Fax: +1.86=
0.760.6014 On Dec 29, 2009, at 9:14 AM, Ethan Wu wrote: Dea=
r all, >   I am fitting a PK/PD model using FOCE INT.  I do see a h=
igh correlation between WRES >   I wonder could someone help me with =
implementation of AR residual error model  in NONMEM. >Many thanks! H=
appy new year to all. > >
Received on Tue Dec 29 2009 - 14:58:07 EST