| RUNPCA | Perform principal component analysis (PCA) using singular value decomposition (SVD) using Matlab svd() or svds() >> inv(eigvec)*data = pc; |
| Usage: | >> [pc,eigvec,sv] = runpca(data); >> [pc,eigvec,sv] = runpca(data,num,norm) |
| Inputs: | |
data |
input data matrix (rows are variables, columns observations) |
num |
number of principal comps to return {def|0|[] -> rows in data} |
norm |
1/0 = do/don't normalize the eigvec's to be equivariant {def|0 -> no normalization} |
| Outputs: | |
pc |
the principal components, i.e. >> inv(eigvec)*data = pc; |
eigvec |
the inverse weight matrix (=eigenvectors). >> data = eigvec*pc; |
sv |
the singular values (=eigenvalues) |
| Author: | Colin Humphries, CNL / Salk Institute, 1997 |
| See also: | runica |