lyngby_sopressvd

 lyngby_sopressvd     - SOP with SVD on the residual

        function [U,S,V] = lyngby_sopresvd(X, R,
            'PropertyName', 'PropertyValue')

       Input:  X   The datamatrix
               R   The run structure
               PropertyName:
                  'Paradigm'         The paradigm 
                  'CanComponents'    { 5 } Number of canonical values
                                     maintained
                  'PrincComponents'  { (all) } Number of principal
                                     components maintained 

       Output: U   Canonical/Eigen sequences
               S   Canonical/Eigen values
               V   Canonical/Eigen images

       Perform Strother Orthonormalized Partial Least Square (SOP) on
       the data matrix X by using the run structure (group structure)
       R. The analysis will only the return number of components
       specified by 'CanComponents'.
       The subspace spanned by the result of this analysis is then
       returned as the first part of the result [U,S,V]. 
       The second part of the result is found by an SVD of the
       residual of the data matrix, ie, an SVD of the space
       orthogonal to SOP space. 

       See also: lyngby_sop_main, lyngby_opls, lyngby_cva, lyngby_svd