# Re: \$ERROR and LOGIT

From: nonmem
Date: Tue, 15 Dec 2009 16:50:42 +0000 (GMT)

Looks genius. Why this simple modification of the code did not cross my mind? Probably because I have worked 80 hr/week.... And did not see this problem before.

----- Original Message -----
From: Leonid Gibiansky
Date: Tuesday, December 15, 2009 11:13 am
Subject: Re: [NMusers] \$ERROR and LOGIT
To: nonmem
Cc: nmusers

> Hi Pavel,
> Sorry, I cannot help with the general solution.
> For your particular case, you can try several transformations
> with
> various but fixed THETA(10) values (prepare transformed DVs
> similar to
> how you would do it with log-transformation), and then fit
> transformed
> variable with additive error model (is this what you need?). You
> will
> not be able to compare OF, but you may experiment with THETA(10)
> to
> select the one that provides the best fit, and then use it.
> Thanks
> Leonid
>
> newDV=LOG((VAS/10+THETA(10))/(1-VAS/10+THETA(10))); should be in
> the
> data file
> ...
> model for VAS
> VPRED= prediction of VAS
> ...
> Z=DLOG((VPRED/10+THETA(10))/(1-VPRED/10+THETA(10)))
> Y= Z + ERR(1)
>
> --------------------------------------
> Leonid Gibiansky, Ph.D.
> President, QuantPharm LLC
> web: www.quantpharm.com
> e-mail: LGibiansky at quantpharm.com
> tel: (301) 767 5566
>
>
>
>
> nonmem
> > Leonid,
> >
> > That is an approximation and approximations can be good. For
> example,
> > if VAS does not approach high values, the exponential model
> can work
> > very well. A modified CV-like model of error (deep
> modifications)
> > supported very reasonable predictions. My proble is not in
> the
> > understanting that a transformation can improve the fit. The
> problem is
> > in implementing it in NONMEM. I have tried about several
> > transformations before I sent the email.
> >
> > Here is the problem: It seems like NONMEM is simlifying the
> error
> > models the same way it simplifies the exponential model of
> error. The
> > question is how to get rid of it. If we cannot use
> transformations,
> > what is the point to use them? My only hope is that some new
> methods
> > implemented in NONMEM7 do not do it.
> > (The abstract provides somewhat limited information on what
> was done.
> > Without aditional information, it is hard to replicate.)
> >
> > Thanks,
> > Pavel
> >
> >
> > ----- Original Message -----
> > From: Leonid Gibiansky
> > Date: Tuesday, December 15, 2009 10:01 am
> > Subject: Re: [NMusers] \$ERROR and LOGIT
> > To: nonmem
> > Cc: nmusers
> >
> > > Pavel,
> > > I do not see any justification for a proportional or exponential
> > > model:
> > > no reasons to believe that error of measurement is proportional
> > > to the
> > > value. I would try simple additive error model. In simulations,
> > > one can
> > > truncate at 0 and 10.
> > > The poster that I mentioned specifically discussed various
> > > approached to
> > > the problem that you are trying to solve.
> > > Thanks
> > > Leonid
> > >
> > >
> > > --------------------------------------
> > > Leonid Gibiansky, Ph.D.
> > > President, QuantPharm LLC
> > > web: www.quantpharm.com
> > > e-mail: LGibiansky at quantpharm.com
> > > tel: (301) 767 5566
> > >
> > >
> > >
> > >
> > > nonmem
> > > > Leonid,
> > > >
> > > > This is about visual analog scale. There are a lot of 0
> and 1
> > > values
> > > > (actually, VAS changes from 0 to 10 in this case, but it can
> > > be divided
> > > > by 10). There are articles, presentatione and dissertations
> > > which use
> > > > logit. So, I try diffrent transformations including logit.
> > > >
> > > > CV error works OK, but I still try to take care of the skewed
> > > > distribution.
> > > >
> > > > When I use exponential error, NONMEM transforms it into CV
> > > error.
> > > > Later, simulations do not make sense because NONMEM does not
> > > do the
> > > > same. Exactly as described in the nonmem6 manual.
> > > >
> > > > Thank you for the article. I'll keep digging.
> > > >
> > > > Pavel
> > > >
> > > >
> > > >
> > > > ----- Original Message -----
> > > > From: Leonid Gibiansky
> > > > Date: Tuesday, December 15, 2009 1:01 am
> > > > Subject: Re: [NMusers] \$ERROR and LOGIT
> > > > To: nonmem
> > > > Cc: nmusers
> > > >
> > > > > Pavel,
> > > > > I am not sure what is the problem with the log-
> > > transformation of
> > > > > the
> > > > > data. log(x) = infinity only if x = infinity, do you have
> > > > > infinite
> > > > > observations in your data set? If not, then transformed data
> > > > > cannot be
> > > > > equal to infinity.
> > > > > log(x) = - infinity only if x=0
> > > > > do you have BQL observations coded as zeros? If so, you
> cannot> > > > use
> > > > > exponential error model. But you can either exclude
> BQLs (and
> > > > > use
> > > > > log-transformation) or treat them as BQLs (and still use
> > > > > log-transformation).
> > > > >
> > > > > Looks like your prediction F is between 0 and 1. I do
> not think
> > > > > that
> > > > > exponential error is appropriate for this type of data. Could
> > > > > you
> > > > > elaborate what exactly you are modeling? If this is indeed
> > > > > interval
> > > > > data, this poster can be relevant (Estimating Transformations
> > > > > for
> > > > > Population Models of Continuous, Closed Interval Data,
> Matthew> > > > M.
> > > > > Hutmacher and Jonathan L. French):
> > > > >
> > > > > http://www.page-meeting.org/default.asp?abstract=1463
> > > > >
> > > > > Thanks
> > > > > Leonid
> > > > >
> > > > > --------------------------------------
> > > > > Leonid Gibiansky, Ph.D.
> > > > > President, QuantPharm LLC
> > > > > web: www.quantpharm.com
> > > > > e-mail: LGibiansky at quantpharm.com
> > > > > tel: (301) 767 5566
> > > > >
> > > > >
> > > > >
> > > > >
> > > > > nonmem
> > > > > >
> > > > > > Hello,
> > > > > >
> > > > > > NONMEM has the following property related to intra-subject
> > > > > variability:>
> > > > > > "During estimation by the first-order method, the
> exponential> > > > model and
> > > > > > proportional models give identical results, i.e., NONMEM
> > > > > cannot
> > > > > > distinguish between them." So, NONMEM transforms
> > > > > F*DEXP(ERR(1)) into F
> > > > > > + F*ERR(1).
> > > > > >
> > > > > > Is there an easy around it? / /I try to code the logit
> > > > > transformation.
> > > > > > I cannot log-transform the original data as it is
> > > suggested in
> > > > > some
> > > > > > publications including the presentation by Plan and
> Karlsson> > > > (Uppsala)
> > > > > > because many values will be equal to plus or minus
> infinity.> > > > Will
> > > > > > NONMEM "linearize" the following code:
> > > > > >
> > > > > > Z = DLOG((F+THETA(10))/(1-F+THETA(10)))
> > > > > > Y = DEXP(Z + ERR(1))/(1 + DEXP(Z + ERR(1)))
> > > > > >
> > > > > >
> > > > > >
> > > > > > Thanks!
> > > > > >
> > > > > > Pavel
> > > > > >
> > > > > >
> > > > > >
> > > > >
> > >
>

Received on Tue Dec 15 2009 - 11:50:42 EST

The NONMEM Users Network is maintained by ICON plc. Requests to subscribe to the network should be sent to: nmusers-request@iconplc.com.

Once subscribed, you may contribute to the discussion by emailing: nmusers@globomaxnm.com.