Department of Statistics Unitmark
Dietrich College of Humanities and Social Sciences

Scan Clustering: A False Discovery Approach

Publication Date

July, 2004

Publication Type

Tech Report

Author(s)

M. Perone Pacifico, C. Genovese, I. Verdinelli, and L. Wasserman

Abstract

We present a method that scans a random field for localized clusters while controling the fraction of false discoveries. We use a kernel density estimator as the test statistic and correct for the bias in this estimator by a method we introduce in this paper. We also show how to combine information across multiple bandwidths while maintaining false discovery control.