SIGNIFICANCE TESTING OF SINGLE CLASS DISCRIMINATION MODELS

Single class discrimination (SCD) has recently been described for the analysis of multivariate embedded data. It is a method for determining informative axes in the data space which promote clustering of the em bedded, or principal, class about the model origin and dispersal of th e non-embedded class. Significance testing of the eigenvalues obtained in a model has been carried out by randomizing the class membership v ector and recalculating the SCD model 500 times. These random simulati ons enable the determination of the permutation distribution under the null hypothesis of no association, and hence can be used to determine the significance of the first eigenvalue. A method is described to es timate the permutation distribution of the second and subsequent eigen values conditional on the fact that the previous eigenvectors in the S CD model have been accepted as significant.