plsPV

plsPV

PLS=plsPV(X,Y,nfold,type,permutations,method)

Calculates a partial least squares regression of Y on X using PLS components of X by SIMPLS algorithm and outputs an array with scores, and loadings for each PLS component and the PLS coefficients vector. This function uses nfold permutation-validation to avoid overfitting, and the user will be asked to select the optimal number of PLS components to avoid over- and under-fitting the data. This function can perform PLS regression or PLS discriminant analysis. Scaling is performed during validation in order to avoid the introduction of bias.

Graphical representation: NA

Known bugs: NA

Typical next steps: VisScores, VisLoadings1D,

also see related function with n-fold cross-validation: plsCV

Tags: Multivariate-Statistics, Group-Separation

References: