The Statistics Seminar spearker for Wednesday, Oct. 12, 2016 will be Kai Zhang, an assistant professor at the Department of Statistics and Operations Research at the University of North Carolina, Chapel Hill. He received the Ph.D. degree in Mathematics from Temple University in 2007 and the Ph.D. degree in Statistics from the Wharton School, University of Pennsylvania in 2012. His research interests include high-dimensional inference, causal inference and observational studies, and robust statistics.
For a full publications list, please visit Zhang's webpage.
Title: BET on Independence
Abstract: We study the problem of model-free dependence detection. This problem can be difficult even when the marginal distributions are known. We explain this difficulty by showing the impossibility to uniformly consistently distinguish degeneracy from independence with one test. To make model-free dependence detection a tractable problem, we introduce the concept of binary expansion statistics (BEStat) and propose the binary expansion testing (BET) framework. Through simple mathematics, we convert the dependence detection problem to a multiple testing problem. Besides being model-free, BET also enjoys many other advantages which include (1) invariance to marginal monotone transformations, (2) clear interpretability of local relationships upon rejection, and (3) close connections to computing for efficient algorithms.
Refreshments will be served following the seminar in 1181 Comstock Hall.