Activity Number:
|
163
- Methods for Complex Data: The Next Generation
|
Type:
|
Topic Contributed
|
Date/Time:
|
Monday, July 29, 2019 : 10:30 AM to 12:20 PM
|
Sponsor:
|
Business and Economic Statistics Section
|
Abstract #301859
|
|
Title:
|
Sequential Change-Point Detection for High-Dimensional and Non-Euclidean Data
|
Author(s):
|
Lynna Chu* and Hao Chen
|
Companies:
|
University of California, Davis and University of California, Davis
|
Keywords:
|
sequential detection;
high-dimensional;
non-Euclidean data;
graph-based tests;
change-point;
non-parametric
|
Abstract:
|
In many modern applications, high-dimensional/non-Euclidean data sequences are collected to study complicated phenomena over time. It is of scientific significance to detect anomaly events as data are being collected. We study a nonparametric framework that utilizes nearest neighbor information among the observations and can be applied to data in arbitrary dimension and non-Euclidean data as long as a similarity measure on the sample space can be defined. We consider new test statistics under this framework that can make more positive detections and can detect anomaly events sooner than the existing test with the false discovery rate controlled at the same level. Analytic formulas for approximating the average run lengths of the new approaches are derived to make them fast applicable to large datasets.
|
Authors who are presenting talks have a * after their name.