Analysis of Asymmetric Dependence in Contingency Tables: Subcopula-Based Regression Approach

• Activity Number: 510 - Recent Development in the Assessment and Modeling of Asymmetric Dependence
• Type: Topic Contributed
• Date/Time: Wednesday, August 1, 2018 : 10:30 AM to 12:20 PM
• Sponsor: IMS
• Author(s): Daeyoung Kim* and Zheng Wei
• Companies: University of Massachusetts Amherst and University of Maine
• Keywords: asymmetric association; contingency table; regression; subcopula

Abstract

For the analysis of a two-way contingency table, a new asymmetric association measure is developed. The proposed method uses the subcopula-based regression between the discrete variables to measure the asymmetric predictive powers of the variables of interest. Unlike the existing measures of asymmetric association, the subcopula-based measure is insensitive to the number of categories in a variable, and thus, the magnitude of the proposed measure can be interpreted as the degree of asymmetric association in the contingency table. The theoretical properties of the proposed subcopula-based asymmetric association measure are investigated. We illustrate the performance and advantages of the proposed measure using simulation studies and real data examples.