Multivariate variographic versus bilinear data modeling

Two contrasting multivariate data sets (a process data series vs. a 1-D geochemical soil profile) are analyzed to illustrate the benefits of using bilinear projection scores for variographic characterization instead of using individual variables. By using absolute variograms on a validated number of component scores, it is possible to make a combined multivariate chemometrics-variogram characterization of heterogeneous processes and materials as well as 1-D transects, no longer restricted to a one-variable-at-a-time framework. The usefulness and information on variographic modeling based on scores are illustrated. A new test for randomness of a variogram is presented.