Selection of a Multivariate Calibration Method

1. Methods to be considered for multivariate calibration Many methods for multivariate calibration have been proposed. It turns out that many of the methods perform similarly. To avoid confusion due to use of many different methods, it is suggested that only the following should be considered: Multiple linear regression (MLR) Principal component regression (PCR) Partial least squares (PLS) Neural networks (NN) Locally weighted regression (LWR) Radial basis functions combined with PLS (RBF-PLS) They have been included on the basis of theoretical considerations, confirmed by an intercomparison of their performances carried out on near infrared data sets with different data structures. Their main advantages and disadvantages are described in the following sections