Boomer Manual and Download
PharmPK Listserv and other PK Resources
Previous Page Course Index Next Page

Data Output

Table of Observed versus Calculated Data

The table of observed data should be check for data entry errors. Also, comparison of observed and calculated data can give a good indication of the final fit from the nonlinear regression analysis.

Boomer
Table of observed and calculated data
WinNonlin
Table of observed and calculated data

Table 9.3.1 Table of observed and calculated data

Notice the similarities between the Calculated and Observed Data indicating a reasonably good fit to the data.

Weight Applied to each Data Point

Boomer
Table of observed data and weight
WinNonlin
Table of observed and calculated data

Table 9.3.2 Table of observed data and SQRT(weight)

You should look for very high or very low weight values which may indicate data entry errors or outliers. Possibly another weighting scheme may be more suitable.

Weighted Residual

Boomer
Table of weighted residuals
WinNonlin
Table of Weighted Residuals

Table 9.3.3 Table of Observed Data and the Weighted Residuals

This table of weighted residuals can be quite instructive. A good fit would be indicated by weighted residuals with similar values and random change of sign. This would indicate that the best fit line was between the observed data points.


return to the Course index


This page was last modified: Sunday, 28th Jul 2024 at 4:48 pm


Privacy Statement - 25 May 2018

iBook and pdf versions of this material and other PK material is available

Copyright © 2001-2022 David W. A. Bourne (david@boomer.org)