Description of the program: svd

This program performs a global SVD. It gives the singular values of the covariance matrix and if the -q flag is set either the data in the basis of the eigenvectors or, if a the parameter of the -q flag is smaller than the embedding dimension, the projected scalar time series. Some authors state such a projection to be a noise reduction scheme. But this is only true if both, the system and the measurement function, are linear.

Usage:

svd [Options]

Everything not being a valid option will be interpreted as a potential datafile name. Given no datafile at all, means read stdin. Also - means stdin

Possible options are:

Option Description Default
-l# number of data to be used whole file
-x# number of lines to be ignored 0
-c# column to be read 1
-m# dimension of the basis 2
-d# delay 1
-q# project down to # dimensions and write the projected time series
or if # equals the embedding dimension, write the vectors in the svd basis 		none (only write eigenvalues)
-o[#] output file name -o without argument means 'datafile'.svd
(or stdin.svd if data were read from stdin)
without -o means stdout
-V# verbosity level
  0: only panic messages
  1: add input/output messages 		1
-h show these options none

Description of the Output:

The output consists of the singular values and if the -q option was used, the vectors in the SVD basis (if the dimension projected down to equalled the embedding dimension), or the projected (filtered) time series (if the dimension projected down to was smaller than the embedding dimension).
View the C-sources.
See also pc
Table of Contents * TISEAN home