|
From: Lindauer, A. (Germany) <And...@sg...> - 2016-05-27 21:16:13
|
Dear PsN-Team and users,
I just came across an issue related to the bootstrap resampling when IGNORE statements are used in the code:
I have a big NONMEM dataset comprised of several studies and treatment arms. I exclude some arms/studies during modeling using IGNORE statements (e.g. placebo group).
Now, when I did a bootstrap, the resulting distributions looked kind of odd ('cauchy'). Then I realized what was going on: Although PsN applies the IGNORE statements correctly in each bootstrap replicate run, the resampling in the beginning is done on the entire dataset. So this means that there may be replicate datasets with a lot of e.g. placebo subjects that are going to be ignored during the run. Not only can this lead to odd distributions, but may also be the cause for crashing/unsuccessful replicate runs.
I know the issue can be partly addressed by using -stratify_on - but this doesn't solve the problem completely.
As a workaround I'm running my bootstraps now on a subset of the original dataset, corresponding to the IGNORE statements used in the code.
Maybe in a future version of PsN the resampling could be done AFTER applying the IGNORE/ACCEPT statements to the dataset ? Or at least mentioning the issue in the documentation would be helpful.
Kind regards, Andreas.
Andreas Lindauer, PhD
Agriculture, Food and Life
Life Science Services - Exprimo
Senior Consultant
Phone:
+49 2159 5323 753
E-mail:
and...@sg...<mailto:and...@sg...>
Web:
www.exprimo.com<http://www.exprimo.com/>
SGS Exprimo NV
Generaal De Wittelaan 19A Bus 5
B-2800 Mechelen, Belgium
Switchboard: +32 15 27 32 45
[cid:image001.jpg@01D1B86A.E6ECD560]<http://www.sgs.com/~/media/Global/Documents/Brochures/SGS%20LSS%20PK%20Modeling%20and%20Simulation%20EN%2014.pdf>
Information in this email and any attachments is confidential and
intended solely for the use of the individual(s) to whom it is addressed
or otherwise directed. Please note that any views or opinions presented
in this email are solely those of the author and do not necessarily
represent those of the Company.
Finally, the recipient should check this email and any attachments for
the presence of viruses. The Company accepts no liability for any damage
caused by any virus transmitted by this email.
All SGS services are rendered in accordance with the applicable SGS
conditions of service available on request and accessible at
http://www.sgs.com/en/Terms-and-Conditions.aspx
|