ABSTRACT
Classical least-squares (CLS), sometimes known as K-matrix calibration, is so called because, originally, it involved the application of multiple linear regression to the classical expression of the Beer-Lambert Law of spectroscopy: A=KC To produce a calibration using classical least-squares, the chapter deals with a training set consisting of a concentration matrix, C, and an absorbance matrix, A, for known calibration samples. The chapter examines how well CLS was able to fit the training set data. If the fit to the training data is generally poor it could be caused by large errors in the expected concentration values as determined by the referee method. Usually, PRESS should be calculated separately for each predicted component , and the calibration optimized individually for each component. There are any number of variations that can be applied to the CLS technique. The chapter also examines the results we get from a CLS calibration with nonzero intercept.
