U-statistical inference for hierarchical clustering

Clustering methods are a valuable tool for the identification of patterns in high dimensional data with applications in many scientific problems. However, quantifying uncertainty in clustering is a challenging problem, particularly when dealing with High Dimension Low Sample Size (HDLSS) data. We develop here a U-statistics based clustering approach that assesses statistical significance in clustering and is specifically tailored to HDLSS scenarios. These non-parametric methods rely on very few assumptions about the data, and thus can be applied to a wide range of datasets for which the euclidean distance captures relevant features. We propose two significance clustering algorithms, a hierarchical method and a non-nested version. In order to do so, we first propose an extension of a relevant U-statistics and develop its asymptotic theory. Our methods are tested through extensive simulations and found to be more powerful than competing alternatives. They are further showcased in two applications ranging from genetics to image recognition problems.