msvd

Location: CuBatch\Model_PCA\private\msvd.m

function [S,V,D,Xpredicted] = msvd (X,fn,center,conv,maxiter);

Description:
svd(X) with missing data (Nipals algorithm)

Inputs:
X: (m x n) - data matrix with m objects and n variables
fn: number of factors
center: center - if center=1 - perform data centering
conv: convergence criterion (optional, default 1e-10)
maxiter: maximal number of iterations (optional, default 1000)

Outputs:
S: left singular vectors
V: diagonal matrix filled with singular values.
D: right singular vectors
Xpredicted: data matrix reconstructed with fn factors

Called by:
acpf

Subroutines:
Internal: None
External: inimiss

Author:
Beata Walczak (chemo toolbox): slightly modified by C. Durville
ChemoAC, V.U.B., Laarbeeklaan 103,
B-1090 Brussel
Belgium

Last modified: 15-Nov-2002 13:04:44

Contact: Christophe Durville, cdurpp@ifrance.com

References