偏最小二乘 非线性 matlab,有做Matlab partial least squares (偏最小二乘的)的吗?

R2009版本的好像就有PLS函数。

function [Xloadings,Yloadings,Xscores,Yscores, ...

beta,pctVar,mse,stats] = plsregress(X,Y,ncomp,varargin)

%PLSREGRESS Partial least squares regression.

%   [XLOADINGS,YLOADINGS] = PLSREGRESS(X,Y,NCOMP) computes a partial least

%   squares regression of Y on X, using NCOMP PLS components or latent

%   factors, and returns the predictor and response loadings.  X is an N-by-P

%   matrix of predictor variables, with rows corresponding to observations,

%   columns to variables.  Y is an N-by-M response matrix.  XLOADINGS is a

%   P-by-NCOMP matrix of predictor loadings, where each row of XLOADINGS

%   contains coefficients that define a linear combination of PLS components

%   that approximate the original predictor variables.  YLOADINGS is an

%   M-by-NCOMP matrix of response loadings, where each row of YLOADINGS

%   contains coefficients that define a linear combination of PLS components

%   that approximate the original response variables.

%

%   [XLOADINGS,YLOADINGS,XSCORES] = PLSREGRESS(X,Y,NCOMP) returns the

%   predictor scores, i.e., the PLS components that are linear combinations of

%   the variables in X.  XSCORES is an N-by-NCOMP orthonormal matrix with rows

%   corresponding to observations, columns to components.

%

%   [XLOADINGS,YLOADINGS,XSCORES,YSCORES] = PLSREGRESS(X,Y,NCOMP)

%   returns the response scores, i.e., the linear combinations of the

%   responses with which the PLS components XSCORES have maximum covariance.

%   YSCORES is an N-by-NCOMP matrix with rows corresponding to observations,

%   columns to components.  YSCORES is neither orthogonal nor normalized.

%

%   PLSREGRESS uses the SIMPLS algorithm, and first centers X and Y by

%   subtracting off column means to get centered variables X0 and Y0.

%   However, it does not rescale the columns.  To perform partial least

%   squares regression with standardized variables, use ZSCORE to normalize X

%   and Y.