From the fundamental parts of PLS‐DA, Fisher's canonical discriminant analysis (FCDA) and Powered PLS (PPLS), we develop the concept of powered PLS for classification problems (PPLS‐DA). By taking advantage of a sequence of data reducing linear transformations (consistent with the computation of ordinary PLS‐DA components), PPLS‐DA computes each component from the transformed data by maximization of a parameterized Rayleigh quotient associated with FCDA. Models found by the powered PLS methodology can contribute to reveal the relevance of particular predictors and often requires fewer and simpler components than their ordinary PLS counterparts. From the possibility of imposing restrictions on the powers available for optimization we obtain an explorative approach to predictive modeling not available to the traditional PLS methods. Copyright © 2008 John Wiley & Sons, Ltd.
[1]
T. Næs,et al.
Multivariate strategies for classification based on NIR-spectra—with application to mayonnaise
,
1999
.
[2]
T. Næs,et al.
From dummy regression to prior probabilities in PLS‐DA
,
2007
.
[3]
M. Barker,et al.
Partial least squares for discrimination
,
2003
.
[4]
Ulf G. Indahl,et al.
A twist to partial least squares regression
,
2005
.
[5]
Hicham Noçairi,et al.
Discrimination on latent components with respect to patterns. Application to multicollinear data
,
2005,
Comput. Stat. Data Anal..