Gram

From Eigenvector Research Documentation Wiki
Revision as of 19:18, 9 October 2008 by imported>Bob
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to navigation Jump to search
The printable version is no longer supported and may have rendering errors. Please update your browser bookmarks and please use the default browser print function instead.

Purpose

Generalized rank annihilation method.

Synopsis

[ord1,ord2,ssq,aeigs,beigs] = gram(a,b,tol,scl1,scl2,out)

Description

GRAM determines the joint invariant subspaces common to the two input matrices a and b, the ratio of their magnitudes ssq, and the response in both modes/orders ord1 and ord2. GRAM assumes that the input matrices a and b are bilinear, i.e. are the summation over outer products.

Inputs are the two response matrices a and b, and the number of factors to calculate or tolerance on the ratio of smallest to largest singular value tol. Optional inputs scl1 and scl2 are scales to plot against when producing plots of the response in each mode/order. Optional input out suppresses plotting and printing of results to the command window when set to 0 {default out = 1}.

Outputs are the pure component responses in each mode ord1 and ord2, the table of eigenvalues and their ratios ssq, and the eigenvalues for each matrix aeigs and beigs.

Inputs

  • a = first input matrix.
  • b = second input matrix.
  • tol = number of factors to calculate or tolerance on the ratio of the samllest to largest singular value.

Optional Inputs

  • scl1 and scl2= scales to plot against.
  • out = input to suppress plotting and printing of results when set to 0 (default value = 1).

Outputs

  • ord1 and ord2 = pure component responses in each mode.
  • ssq = table of eigenvalues and their ratios.
  • aeigs = eigenvalues for a.
  • beigs = eigenvalues for b.

See Also

mpca, parafac, parafac2, tld