A Procedure to Detect General Association Based on Distance of Ranks

• Activity Number: 187
• Type: Contributed
• Date/Time: Monday, August 4, 2014 : 10:30 AM to 12:20 PM
• Sponsor: ENAR
• Author(s): Pratyaydipta Rudra*+ and Fred Wright
• Companies: University of North Carolina at Chapel Hill and North Carolina State University
• Keywords: Test of independence ; General association ; Rank-based test

Abstract

In modern high-throughput applications, it is often important to identify pairwise associations between variables, and it is desirable to use methods that are powerful and sensitive to different types of associations. We describe a new non-parametric association test for association between two variables that is based on distances of ranks, measuring the concentration of paired points in a scatterplot. Here 'concentration' is quantified using disc-covering approaches that have been employed in analysis of spatial data. Analysis of simulated datasets demonstrate that our method is powerful and robust in comparison to competing general association tests. A variety of real datasets, ranging from studies of cell cycle effects in gene expression to studies involving circular interference transmittance show that the approach provides useful and interpretable results.