Residuals in multiple regression analysis

An examination of residuals can be used as a technique to determine the best model for use in multiple-regression analysis. In addition to R2 and the Fisher F ratio, a complete analysis of residuals has been used by statisticians for many years to estimate bias and to choose the best equation to fit physicochemical as well as biological and psychosociological data. However, residual analysis has seldom appeared in studies in pharmaceutics, pharmacology, or medicinal chemistry. This report gives two examples to show the usefulness of residual analysis in the pharmaceutical sciences.