|
Activity Number:
|
544
- Dynamic Graphical Models and Networks with Applications
|
|
Type:
|
Invited
|
|
Date/Time:
|
Wednesday, July 31, 2019 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
International Indian Statistical Association
|
|
Abstract #300401
|
|
|
Title:
|
On the CUSUM Changepoint Estimator for Network Data
|
|
Author(s):
|
Shirshendu Chatterjee* and Sharmodeep Bhattacharyya and Peter J Bickel and Soumendu Sundar Mukherjee
|
|
Companies:
|
City University of New York, City College and Oregon State University and University of California, Berkeley and Indian statistical Institute
|
|
Keywords:
|
dynamic network;
change point detection;
network inference;
time series
|
|
Abstract:
|
In this article, we consider the problem of changepoint detection in network valued time series. We consider the "offline" case, where the entire time series is observed beforehand. We analyze an appropriate modification of the CUSUM statistic in this problem and obtain rates of convergence of the resulting changepoint estimator involving the sizes of the networks, their sparsity and the number of networks in the series.
|
Authors who are presenting talks have a * after their name.
