Wilks’ Theorems in Some Random Graph Models

Loading Events

Instructors/Speakers

Prof. Ting YAN
Professor
Huazhong Normal University
China

Abstract

In this talk, we will show Wilks’ theorems in two random graph models—the beta model and the Bradley-Terry model. For fixed dimensional parameter testing problems, we show that minus twice log-likelihood ratio statistics in the null converge to the Chi-square distribution although the number of nuisance parameters goes to infinity. For increasing dimensional parameter testing problem, we reveal high dimensional Wilks phenomenon.

Biography

Prof. Ting YAN obtained his Ph.D. degree in University of Science and Technology of China in 2012, works as a full professor at Huazhong Normal University. His main research interests include Network models, Random Graphs, Paired comparisons, etc. His research has been supported by three projects of National Natural Science Foundation of China, and he has published more than 30 journal papers, including several top journals in Statistics, such as Annals of Statistics, Biometrika and Journal of the American Statistical Association.

Go to Top