Back to the Top
Dear all,
Does anyone use POP3CM by the RFKA of the University of Washington
[http://tanami.rfpk.washington.edu/software.html - db] to analyze
the population PK and how should the quality of the result be
evaluated in term of the mean and covariance or is there any other
standard to use for the evaluation?
Thank you all.
Zhao Wang, M.D
Northwestern University
Medical School
Anesthesia Research
z-wang.-a-.nwu.edu
Back to the Top
To: Zhao Wang, M.D
From: Bradley Bell, Mathematician
Subject: Statistics and Pop3cm
I wrote Pop3cm as a combination of a prototype population algorithm
together with a prototype interface. I am now working on the actual
product and do not have the time to maintain the Pop3cm prototype.
There is a simulation option in Pop3cm. It was intended that you would
obtain your statistics by simulating different data sets and noting how
much your estimates varied.
The question that you ask is a very good one in general. I am working on a
method that will allow you to obtain statistics for any population
estimates. One benefit is that it would allow you could compare the
different methods and see how well they approximate the minimizer of the
true marginal likelihood.
In short, what you are asking for is coming, but it will take some time.
You can find a description of some of my recent research in this area at
the web page:
http://www.seanet.com/~bradbell/pub.htm
PharmPK Discussion List Archive Index page
Copyright 1995-2010 David W. A. Bourne (david@boomer.org)