SPCA (Sparse PCA) estimates a PCA-like model with sparsity induced on scores or loadings via L1 norm bounding. Provides NIPASLS (deflation-based) and ASLS (simultaneous) algorithmic options with soft thresholding. The penalty parameter lambda drives selected loadings to zero, improving interpretability of high-dimensional models through automated variable selection.
Morten Arendt Rasmussen [1]Rasmus Bro [1]
[1]Department of Food Science, University of Copenhagen, Denmark
