A0282
Title: Optimal nonparametric inference on network effects with dependent edges
Authors: Wen Zhou - New York University (United States) [presenting]
Yuan Zhang - The Ohio State University (United States)
Wenqin Du - University of Southern California (United States)
Abstract: Testing network effects in weighted directed networks is a foundational problem in econometrics, sociology, and psychology. Yet, the prevalent edge dependency poses a significant methodological challenge. Most existing methods are model-based and come with stringent assumptions, limiting their applicability. In response, a novel, fully nonparametric framework is introduced that requires only minimal regularity assumptions. While inspired by recent developments in U-statistic literature, the approach notably broadens their scopes. Specifically, the challenge of indeterminate degeneracy in the test statistics is identified and carefully addressed, a problem that the aforementioned tools do not handle. Berry-Esseen type bounds are established for the accuracy of type-I error rate control. With the original analysis, the minimax power optimality of the test is also proved. Simulations underscore the superiority of the method in computation speed, accuracy, and numerical robustness compared to competing methods. The method is also applied to the U.S. faculty hiring network data and discovered intriguing findings.
