Back to the Top
Dear all,
In the NCA module of Phoenix, one of the methods for the estimation of =
lambda z is through the Best Fit, and the software would automatically =
select at least 3 data from the last observation. My question is, what's =
the judging criteria used to determine, let's say, that the last 4 =
points is better than the last 3 points? Is it R2, or some other =
parameters calculated? Thanks for your help.
Regards
--=20
Xinting=
Back to the Top
Dear all,
In the NCA module of Phoenix, one of the methods for the estimation of
lambda z is through the Best Fit, and the software would automatically
select at least 3 data from the last observation. My question is, what's
the judging criteria used to determine, let's say, that the last 4
points is better than the last 3 points? Is it R2, or some other
parameters calculated? Thanks for your help.
Regards
Xinting
Back to the Top
Hi Xingting,
In fact, the clarification on the lambda Z estimation was specified
detailedly in the Phoenix User's guide. I cited some important info on
it in the follow:
"WinNonlin(R) estimates lambda-z using the regression with the largest =
adjusted
R2 and:
" If the adjusted R2 does not improve, but is within 0.0001 of the
largest adjusted R2 value, the regression with the larger number of
points is used.
" lambda-z must be positive, and calculated from at least three data =
points."
Hope this can answer your question.
Thanks
MJ
Back to the Top
[This and the next few may be repeats or just delayed - I found another =
folder in my mail client - db]
The following message was posted to: PharmPK
Dear Xinting,
Essentially, the regression is selected based on Rsq and number of data
points.
I would recommend running the automatic fit and then amend the fits
following visual (i.e. subjective) inspection.
Charlie
KinetAssist Ltd.=
Back to the Top
The following message was posted to: PharmPK
Xinting,
Phoenix uses an algorithm that maximizes the adjusted r^2 with 3 or more
data points to select the best-fit regression line. The adjusted r^2
statistic includes a slight penalty for including additional data
points. This tends to minimize the number of data points included in the
calculation while maximizing the fit to the data. The adjusted r^2 is
less than or equal to the r^2 in all cases.
Nathan
Founder and President, PK/PD Associates
nathan.at.learnpkpd.com
Back to the Top
The following message was posted to: PharmPK
Hello Xinting,
In the best fit algorithm of Winnonlin, the selection criteria is
based on Adj R square and Rsquare the combination of points which
would yield Best Fit R square value would be selected by WNL algorithm
Hope Iam clear you can always check the manual of WNL for help
"WinNonlin(R) estimates \0x03BBz using the regression with the largest
adjusted R2"
Regards
Kaushik
Back to the Top
Xinting:
They use the adjusted R2. See attachment - [page 276 Phoenix User's Guide - db]
Chris
Christopher J. Kemper, Ph.D.
Pharma Navigators, LLC
DMPK/Bioanalytical Consulting
Alliance Management/Business Analysis
Want to post a follow-up message on this topic?
If this link does not work with your browser send a follow-up message to PharmPK@boomer.org with "Algorithm of Best Fit in Phoenix WinNonLin" as the subject | Support PharmPK by using the |
Copyright 1995-2011 David W. A. Bourne (david@boomer.org)