Scan Clustering: A False Discovery Approach

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


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.

Heidi Sestrich 2004-07-01
Here is the full PDF text for this technical report. It is 1012431 bytes long.