[Master Index] [Index for export/lyngby]

lyngby_fir_regmatrix

(export/lyngby/lyngby_fir_regmatrix.m)


Function Synopsis

R = lyngby_fir_regmatrix(filterOrder, varargin)

Help text

 lyngby_fir_regmatrix - Construct regularization matrix

       R = lyngby_fir_regmatrix('PropertyName', 'PropertyValue')

       Input:    filterOrder Number of stick functions

       Property: Reg         [ {0} | non-negative ] Regularization
                             strength          
                 RegMethod   [ {smooth} | ridge ] Regularization
                             method 

       Output:   R           Regularization matrix

       This function produces a regularization matrix that can be
       used in a biased estimate. It can be used in the multivariate
       regression model (Y = X*B+U) for estimatrion of B
            
          B = (X'*X + R)^(-1) * X'*Y

       The regularization matrix correspond to eith the 'smooth' FIR
       filter or the ridge regularized FIR filter depending on the
       setting of the 'RegMethod' property.

       See also LYNGBY, LYNGBY_FIR_MAIN, LYNGBY_FIR_TEST_CHI2. 

 $Id: lyngby_fir_regmatrix.m,v 1.2 2003/02/21 14:27:05 fnielsen Exp $

Cross-Reference Information

This function calls

Produced by mat2html on Wed Jul 29 15:43:40 2009
Cross-Directory links are: OFF