Research article
An evaluation of R2 as an inadequate measure for nonlinear models in pharmacological and biochemical research: a Monte Carlo approach
-
* Corresponding author: Andrej-Nikolai Spiess a.spiess@uke.uni-hamburg.de
1 Department of Andrology, University Hospital Hamburg-Eppendorf, Hamburg, Germany
2 Department of Mathematics, University of Hamburg, Hamburg, Germany
BMC Pharmacology 2010, 10:6 doi:10.1186/1471-2210-10-6
Published: 7 June 2010Additional files
Additional File 1:
Mathematical derivation and concise discussion of features and pitfalls in the use of R2 in nonlinear regression and description of the simulation setup.
Format: PDF Size: 53KB Download file
This file can be viewed with: Adobe Acrobat Reader
Additional File 2:
R code used for conducting the simulations. 'pcrsim' of package 'qpcR' is the workhorse function that creates simulated data starting from the fitted value, adding a desired noise structure and testing different sigmoidal models on the perturbed data. 'code' collects the results and summarizes the data as shown in this manuscript. R script file for the R statistical environment.
Format: ZIP Size: 1KB Download file