Partial least squares discriminant analysis: taking the magic away

Short summary

Partial Least Squares Discriminant Analysis (PLS-DA) offers no significant advantages over traditional classification methods like Euclidean distance to centroids or linear discriminant analysis for equal class sizes, and is prone to pitfalls with unequal or multiple classes, overfitting, and score plots.

AI-generated from the title and abstract; the full text is not read.