PLS_Toolbox Documentation: nippls	< ncrossval	normaliz >

nippls

Purpose

NIPALS Partial Least Squares computational engine.

Synopsis

[reg,ssq,xlds,ylds,wts,xscrs,yscrs,bin] = nippls(x,y,ncomp,options)

options = nippls('options')

Description

Performs PLS regression using NIPALS algorithm.

INPUTS:

                         x =   X-block (M by Nx) and

                         y =   Y-block (M by Ny).

OPTIONAL INPUTS:

               nocomp =   number of components {default = rank of X-block}, and

      options =   discussed below.

The default options can be retreived using: options = nippls('options');.

OUTPUTS:

                     reg =   matrix of regression vectors,

                     ssq =   the sum of squares captured (ssq),

                   xlds =   X-block loadings,

                   ylds =   Y-block loadings,

                     wts =   X-block weights,

                 xscrs =   X-block scores,

                 yscrs =   Y-block scores, and

                     bin =   the inner relation coefficients.

Note: The regression matrices are ordered in reg such that each Ny (number of y variables) rows correspond to the regression matrix for that particular number of latent variables.

Options

             options =   a structure containing the fields:

                     name:   'options', name indicating that this is an options structure, and

               display:   [ 'off' |{'on'}], governs display to command window.

See Also

analysis, pls, plsnipal, simpls