Re: [NMusers] OFV or Diagnostic Plot ?? Which one rules...

From: Leonid Gibiansky <lgibiansky_at_quantpharm.com>
Date: Wed, 13 Feb 2019 21:34:02 +1300

Sumeet,
DV vs IPRED is only one, and the least helpful plot. You may want to look on=
 DV vs PRED, both in original scale and on log log scale, CWRES vs time, PRE=
D, distributions and correlation of random effects, etc. and only then one c=
an decide which of the models is better. Based on the description, I would g=
uess that model with proportional error provides better fit at very low conc=
entrations, visible in log scale plots. So you may also factor this in in th=
e decision process. If max concentrations are more important, additive error=
 may help but if low concentrations are more important, you may want to use c=
ombined or proportional error.
Regards,
Leonid


> On Feb 13, 2019, at 7:28 PM, Singla, Sumeet K <sumeet-singla_at_uiowa.edu> wr=
ote:
>
> Hi Everyone,
>
> I am fitting two compartment PK model to Marijuana (THC) concentrations. W=
hen I apply proportional error (or proportional plus additive) residual mode=
l, I get pretty good fits (except 15% of subjects) at all time points.
> However, when I apply only additive error residual model, I get perfect fi=
ts in all subjects but objective functional value is increased by about 20 u=
nits. DV vs IPRED reveal all concentrations on line of unity.
> My question is: should I go with additive error model which gives me perfe=
ct fit but higher OFV or should I go with proportional error model which giv=
es me lower OFV but not so good fit in couple of subjects?
>
> Regards,
> Sumeet Singla
> Graduate Student
> Dpt. of Pharmaceutics & Translational Therapeutics
> College of Pharmacy- University of Iowa
>

Received on Wed Feb 13 2019 - 03:34:02 EST

This archive was generated by hypermail 2.3.0 : Fri Sep 27 2019 - 17:02:58 EDT