Seminars

General Departmental Seminar Series

A Weighted Average Likelihood Ratio Test for Clustering of Disease

Ronald Gangnon, Department of Biostatistics & Medical Informatics, University of Wisconsin

Friday, Oct 27, 2000, 1200-100 pm

G5/164 Clinical Sciences Center, 600 Highland Avenue

ABSTRACT

Numerous statistical methods have been proposed for detecting spatial (or spatial-temporal) clustering of disease cases. We consider methods for detecting so-called localized clusters. We propose a new test statistic, the weighted average likelihood ratio (WALR) test, as an alternative to the spatial scan (maximum likelihood ratio) test statistic (Kulldorff and Nagarwalla, 1994; Kulldorff, 1997). We propose an unbiasedness criterion for cluster detection methods and evaluate the bias of the tests through simulation. We also examine the power of the tests through simulations and apply the methods to the well-known New York leukemia data.

Back to General Departmental Seminar Series