Detecting spatial clusters in functional data: New scan statistic approaches

Abstract

We have developed a scan statistic for detecting clusters of functional data indexed in space. The method is based on a distribution-free spatial scan statistic for univariate data. In a simulation study, it always performed better than a nonparametric functional scan statistic. The proposed method can detect smaller spatial clusters than the nonparametric method. Lastly, we used our scan statistic for functional data to search for spatial clusters of abnormal unemployment rates in France over the period 1998–2013 (divided into quarters).