论文信息 - Deflation in multiblock PLS

Deflation in multiblock PLS

This paper describes some of the deflation problems in multiblock PLS. Deflation of X using block scores leads to inferior prediction of Y. Deflation of X using super scores gives the same predictions as standard PLS with all variables in one large X‐block, but the information of the separate blocks gets mixed up owing to the deflation of X. If Y is deflated using the super score instead, these problems disappear. Copyright © 2001 John Wiley & Sons, Ltd.

A. Smilde | J. Westerhuis

[1] Bruce R. Kowalski,et al. Partial least-squares path modelling with latent variables , 1979 .

[2] Bruce R. Kowalski,et al. Prediction of Product Quality from Spectral Data Using the Partial Least-Squares Method , 1984 .

[3] Bruce R. Kowalski,et al. A multivariate method for relating groups of measurements connected by a causal pathway , 1985 .

[4] P. Geladi. Notes on the history and nature of partial least squares (PLS) modelling , 1988 .

[5] L. E. Wangen,et al. A multiblock partial least squares algorithm for investigating complex chemical systems , 1989 .

[6] John F. MacGregor,et al. Process monitoring and diagnosis by multiblock PLS methods , 1994 .

[7] J. A. Westerhuis,et al. Optimisation of the composition and production of mannitol/microcrystalline cellulose tablets , 1996 .

[8] Johan A. Westerhuis,et al. Multivariate modelling of the tablet manufacturing process with wet granulation for tablet optimization and in-process control , 1997 .

[9] Bhupinder S. Dayal,et al. Improved PLS algorithms , 1997 .

[10] Age K. Smilde,et al. Comments on multilinear PLS , 1997 .

[11] J. Westerhuis,et al. Multivariate modelling of the pharmaceutical two‐step process of wet granulation and tableting with multiblock partial least squares , 1997 .

[12] J. Macgregor,et al. Analysis of multiblock and hierarchical PCA and PLS models , 1998 .