ANOVA-TP

ANOVA-Target projection (ANOVA-TP)

ANOVA-Target Projection (ANOVA-TP) is an alternative to the ASCA method, which is also based on the decomposing the experimental data matrix according to a multiple ANOVA scheme. To evaluate and interpret the effect of each design term, the method makes use of PLS followed by target projection (TP), which allows a one-factor representation of the model for each column in the design dummy matrix. PLS application follows after proper deflation of the experimental matrix, i.e., to what are called the residuals under the reduced ANOVA model.

References:

F. Marini, D. de Beer, E. Joubert, B. Walczak, Analysis of variance of designed chromatographic data sets: The Analysis of Variance-Target Projection approach, J. Chromatogr. A 1405 (2015) 94-102.

F. Marini, D. de Beer, N.A. Walters, A. de Villiers, E. Joubert, B. Walczak, Multivariate analysis of variance of designed chromatographic data. A case study involving fermentation of rooibos tea, J. Chromatogr. A 1489 (2017) 115-125.

F. Marini, B. Walczak, ANOVA-Target projection (ANOVA-TP). In: S.D. Brown, R. Tauler, B. Walczak (Eds.), Comprehensive Chemometrics, 2nd edition, Elsevier, Amsterdam, 2020, in press.

Download the Matlab functions (and demo) for ANOVA-TP modeling