pcacomp

PURPOSE

SYNOPSIS

DESCRIPTION

Returns the principal components of (X) specified in vector (index)
[COEFF,SCORE,LATENT] = PCACOMP(X,INDEX)

 Inputs:
    (X): Data matrix
    (INDEX): Vector containing indices of principal component vectors required

 Output:
    (COEFF):  [pca(X,'NumComponents',index(1)), pca(X,'NumComponents',index(2)), ... ]
    (SCORE):  Representation of X in the principal component space
    (LATENT): Eigenvalues of COV(X'*X)

CROSS-REFERENCE INFORMATION

• depddp density-enhanced Principal Direction Divisive Partitioning algorithm
• drsc Dimensionality Reduction for Spectral Clustering
• ldakmeans LDA-K-means algorithm
• mddc Minimum Density Divisive Clustering
• ncutdc Minimum Normalised Cut Divisive Clustering
• bisKmeansPP Projection pursuit function template to implement Bisecting K-Means
• depddppp Projection Pursuit function for dePDDP algorithm
• gsep Generic binary cluster separator class
• hp Class implementing generic hyperplane interface
• pddppp Projection Pursuit function for PDDP algorithm
• mdh Minimum Density Hyperplane
• mdhp MDHP implements Minimum Density Hyperplane (inherits from HP class)
• mdpp Minimum Density Projection Pursuit (MDPP) algorithm
• nchp Minimum Normalised Cut Hyperplane (inherits from HP class)
• ncut_sigma Default scaling parameter employed by Gaussian kernel in minimum normalised cut projection pursuit
• ncuth Minimum Normalised Cut Hyperplane
• ncutpp Minimum normalised cut projection pursuit
• schp Class implementing a linear projection subspace of arbitrary dimensions estimated through SCPP
• scpp Spectral Clustering Projection Pursuit (SCPP) (divisive clustering is implemented in scppdc.m)