Keywords - Function groups - @ A B C D E F G H I J K L M N O P Q R S T U V W X Y Z

Library: xplore
See also: glm redun makedesign

Macro: dpls
Description: calculating latent variables, weights, loadings and path coefficients with dynamic partial least squares algorithm

Usage: {w,b,sk,skl,lk,iter} = dpls(w,d,dy,dl,y,genau)
Input:
w a matrix with start weights same dimensions as dy
d a kxk matrix with inner unlagged designs (0 or 1) no diagonal values allowed
dy a lxk matrix with outer designs (0 or 1) rows are counting manifest variables
dl a kxk matrix with inner lagged designs (0 or 1) diagonal elements are showing autoregression
y a nxl matrix with manifest variables (indicators)
genau a scalar with canceling criterion
Output:
wg a matrix with weights
b a matrix with loadings
sk a matrix with path coefficients with dimensions like d (kxk)
skl a matrix with lagged path coefficients with dimensions like d (kxk) and ordered like designed
lk a matrix with latent variables
iter a scalar shows how many iterations used

Example:
library("xplore")
y=read("indikat.dat")    
dy=read("dy.dat")
d=read("d.dat")
dl=read("dl.dat")
w=dy
{wg,b,sk,skl,lk,iter}=dpls(w,d,dy,dl,y,3)
Result:


Library: xplore
See also: glm redun makedesign

Keywords - Function groups - @ A B C D E F G H I J K L M N O P Q R S T U V W X Y Z

Author: Prof.Dr.H.G.Strohe Universitaet Potsdam; programmed by cand rer oec Frank Geppert; See for details: "DPLS Algorithmus und Computerprogramm fuer dynamische Partial-Least-Squares-Modelle"
(C) MD*TECH Method and Data Technologies, 28.6.1999