|
Activity Number:
|
152
- Frontiers of High-Dimensional and Complex Data analysis
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Monday, July 30, 2018 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
International Chinese Statistical Association
|
|
Abstract #328935
|
Presentation
|
|
Title:
|
A Bernstein-Type Inequality for U-Statistics Under Mixing Conditions
|
|
Author(s):
|
Fang Han* and Yandi Shen and Daniela Witten
|
|
Companies:
|
University of Washington and University of Washington and University of Washington
|
|
Keywords:
|
U-statistics;
Cramer-type moderate deviation;
mixing condition;
stochastic regression;
stationarity test
|
|
Abstract:
|
This talk shows a Bernstein-type inequality along with a Cramer-type moderate deviation theorem for non-degenerate U-statistics under alpha- and tau-mixing conditions. The result confirms a conjecture raised by Borisov and Volodko (2015), and is applicable to a wide range of kernel functions via multiple Fourier series expansion. Two statistical applications of our theory are provided, covering estimation and testing problems in high dimensions.
|
Authors who are presenting talks have a * after their name.
