Re: Linear VS LTBS
Hi, my approach has been to use both LTBS and un-transformed data and then
see which one characterizes the data better. Then change initial estimates
and see how the model predicts.
My previous experience was when using untransformed with INTER the model was
not able to always converge specially when the PK is multiphasic. However,
when using LTBS (exponential only) the models converged and predicted data
(Phase1, 2 or 3) quite well.
The current data set I am working with did not follow the exact trend and
that is why I had posed the original question.
Also when performing a preliminary bootstrap with 500 subjects I noticed
that LTBS showed bi-modality in Vc and Ka but not when using the
untransformed data.
Neil
Quoted reply history
On Sun, Aug 23, 2009 at 11:46 PM, Stephen Duffull <
[email protected]> wrote:
> Mats
>
> > I think you're missing an important point. As I wrote to Nick, you will
> > never get concentrations reported regardless of their value. At some
> > point,
> > you will only get the information that concentration is below a limit
> > (LOQ,LOD,LO?). This you should take into account in your design. Error
> > models for concentrations below LO? are not entirely unimportant, but
> > will
> > not have the properties you mention below.
>
> I am happy with either accounting for censoring, or including an additive
> error model or both for optimal design use with proportional error models.
> I don't think that proportional only error models in the absence of the
> above is good. So I believe we agree here.
>
> Steve
> --
>
>
--
Indranil Bhattacharya