On model selection for standard curve in assay development

This paper discusses the selection of an appropriate statistical model for representing standard curve in assay development. This is an important issue in assay validation because the accuracy and reliability of the assay result depend on the selected standard curve. In this study, we propose a selection procedure, which is based on the R2 and the mean squared error of the estimation sample, to determine the 'best' model. An example concerning an assay validation study is used to illustrate the application of the proposed procedure to discriminate among the five commonly used statistical models.

Duke Scholars

Author Shein-Chung Chow Biostatistics & Bioinformatics, Division of Biostatistics