How to Select the Appropriate Degrees of Freedom for Multivariate Calibration

This column addresses the issue of degrees of freedom (df) for regression models. It seems there is some confusion about the use of df for the various calibration and prediction situations the standard error parameters should be comparable and are related to the total independent samples, data channels containing information (that is, wavelengths or wavenumbers), and number of factors or terms in the regression. By convention everyone could just choose a definition, but there is a more correct one that should be verified and discussed for each case. The problem is computing the standard deviation using different degrees of freedom without a more rigorous explanation and then putting so much emphasis on the actual number derived for the standard error of the estimate (SEE) and the standard error of cross validation (SECV), rather than on the computed confidence intervals.