Re: Nonparametric method
If it is indeed similar, one can try to solve is with the similar trick, forcing all etas to work for all subjects:
http://www.metrumrg.com/publications/aaps_2004_etabug.pdf
Leonid
Brian M. Sadler wrote:
> Dear Tom,
>
> How does it differ from the parametric problem of non-influential correlated etas addressed by Stuart in article 99nov142002.html ( http://www.cognigencorp.com/nonmem/nm/99nov142002.html )?
>
> Cheers… Brian
>
> * From: * [EMAIL PROTECTED] [ mailto:[EMAIL PROTECTED] *On Behalf Of *Ludden, Thomas (MYD)
>
> *Sent:* Tuesday, July 24, 2007 12:02 PM
> *To:* [email protected]
> *Subject:* [NMusers] Nonparametric method
>
> Dear nmusers,
>
> Since Rik Schoemaker's original e-mail a few weeks ago, the code associated with the NONPARAMETRIC option has been reviewed. There are problems (bugs) that may reveal themselves when the data set contains individuals with different active etas. For example, a data set with some individuals receiving only IV doses while others receive only oral doses may exhibit this bug. For the individuals receiving only IV doses, the eta's associated with absorption-related parameter such as F1, ALAG1, KA will be nonactive. At this time it appears, but it is not certain, that the bugs will only affect the resulting estimates when the data set contains individuals with different active etas. Code changes to address these bugs are being reviewed and tested. Anyone who has a control stream and data set for which they have applied the NONPARAMETRIC option is welcome to send these to me ([EMAIL PROTECTED] < mailto:[EMAIL PROTECTED]>) for testing with the modified routine. We hope to provide fixes in the near future.
>
> Tom Ludden
>
> * From: * [EMAIL PROTECTED] [ mailto:[EMAIL PROTECTED] *On Behalf Of *Rik Schoemaker
>
> *Sent:* Wednesday, July 04, 2007 11:02 AM
> *To:* [email protected]
> *Subject:* Re: [NMusers] Nonparametric Modelling
>
> Dear Julie,
>
> Thank you for your comments! Unfortunately the parametric step runs just fine... :-)
>
> Cheers,
>
> Rik
>
> > >> "Julie ANTIC" <[EMAIL PROTECTED]> 04-07-07 3:48 PM >>>
>
> The nonparametric option will only estimate the distribution of ETA. The fixed parameters of your model (THETA(1), THETA(2),...,THETA(8)) should be estimated before the non parametric step with a parametric procedure ($EST...).
>
> By default, if the (parametric) estimation ($EST) step failed, the nonparametric step is not run. I think, it is your problem because the UNCONDITIONAL option just permits to run the nonparametric step when the $EST step has failed.
>
> If I were you, I would try to run the command file without the $NONPARAMETRIC option, and correct eventuel errors...
>
> I hope it'll help, good luck,
>
> Julie Antic
>
Quoted reply history
> ----- Original Message -----
>
> * From: * Rik Schoemaker <mailto:[EMAIL PROTECTED]>
>
> * To: * [email protected] <mailto:[email protected]>
>
> * Sent: * Wednesday, July 04, 2007 2:30 PM
>
> * Subject: * [NMusers] Nonparametric Modelling
>
> Dear all,
>
> I'm trying to implement nonparametric population parameter
> estimation in NONMEM VI but I run into the following error message:
>
> NONPARAMETRIC ESTIMATE REQUESTED. BUT NO INDIVIDUAL RECORDS EXIST
> WITH DATA MODELED WITH ACTIVE ETA'S
>
> As a result no tables are output at all. However, if I leave out the
>
> $NONP UNCONDITIONAL
> the model runs fine and I do get my tables...
>
> Any clues?
>
> Thanks in advance!
>
> Rik Schoemaker
>
> $ PROB PK C1
> $INPUT ID TRT TIME NDV DV CMT AMT SS RATE II MDV WT
> $DATA DRUG4.NM
> $SUBROUTINES ADVAN4 TRANS4
> $PK
> TABS = THETA(1)*EXP(ETA(1))
> KA = 0.692/TABS
> S2 = THETA(2)*EXP(ETA(2))
> V2 = S2
> V3 = THETA(3)*EXP(ETA(3))
> CL = THETA(4)*EXP(ETA(4))
> Q = THETA(5)*EXP(ETA(5))
> ALAG1 = THETA(6)*EXP(ETA(6))
> F1 = THETA(7)*EXP(ETA(7))
> $ERROR
> IPRED = LOG(.025)
> W = THETA(8)
> IF(F.GT.0) IPRED = LOG(F)
> IRES = IPRED-DV
> IWRES = IRES/W
> Y = IPRED+ERR(1)*W
> $THETA 1 15 60 12 15 0.6 0.45 0.3
> $OMEGA .1 .1 .1 .1 .1 .1 .1
> $SIGMA 1 FIXED
> $EST PRINT=1 MAXEVAL 9999 POSTHOC NOABORT METHOD=1
> MSFO=msfb1
> $NONP UNCONDITIONAL
> $COV PRINT=E
> $TABLE ID TIME NDV DV CMT AMT SS RATE II MDV WT
> IPRED IRES IWRES FILE = SDTAB1
> ONEHEADER NOPRINT
> $TABLE ID ALAG1 TABS S2 V3 CL Q F1 FILE=PATAB1
> ONEHEADER NOPRINT
>
> First few data lines:
>
> 1 0 0 . . 1 175000 1 . 12 1 58
> 1 0 0.05 . . 2 . . . . 1 58
> 1 0 0.2166666667 . . 2 . . . . 1 58
> 1 0 0.3833333333 . . 2 . . . . 1 58
> 1 0 0.55 . . 2 . . . . 1 58
> 1 0 0.7166666667 . . 2 . . . . 1 58
> 1 0 0.8833333333 . . 2 . . . . 1 58
> 1 0 1 1206 7.0950643773 2 . . . . 0 58
> 1 0 1.05 . . 2 . . . . 1 58
> 1 0 1.2166666667 . . 2 . . . . 1 58
> 1 0 1.3833333333 . . 2 . . . . 1 58
> 1 0 1.55 . . 2 . . . . 1 58
> 1 0 1.7166666667 . . 2 . . . . 1 58
> 1 0 1.8833333333 . . 2 . . . . 1 58
> 1 0 2 1206 7.0950643773 2 . . . . 0 58
> 1 0 2.05 . . 2 . . . . 1 58
> 1 0 2.2166666667 . . 2 . . . . 1 58
> 1 0 2.3833333333 . . 2 . . . . 1 58
> 1 0 2.55 . . 2 . . . . 1 58
> 1 0 2.7166666667 . . 2 . . . . 1 58
> 1 0 2.8833333333 . . 2 . . . . 1 58
> 1 0 3 975 6.882437471 2 . . . . 0 58
>
> _______________________________________________________________________________________________________________________________