A Sparse PCA Approach to Clustering

Abstract

We discuss a clustering method for Gaussian mixture model based on the sparse principal component analysis (SPCA) method and compare it with the IF-PCA method. We also discuss the dependent case where the covariance matrix $\hat{I}\pounds$ is not necessarily diagonal.

This paper is part of a discussion of the paper "Important feature PCA for high dimensional clustering" by Jiashun Jin and Wanjie Wang to appear in The Annals of Statistics