Activity Number:
|
15
- Networks, Multivariate Analysis, and Time Series
|
Type:
|
Topic Contributed
|
Date/Time:
|
Sunday, July 30, 2017 : 2:00 PM to 3:50 PM
|
Sponsor:
|
Royal Statistical Society
|
Abstract #322699
|
View Presentation
|
Title:
|
On Aggregation of Strongly Dependent Network Flows
|
Author(s):
|
Jan Beran and Sucharita Ghosh and Haiyan Liu*
|
Companies:
|
University of Konstanz and Swiss Federal Research Institute WSL and University of Leeds
|
Keywords:
|
routing matrix ;
OD-flow ;
link count time series ;
long-range dependence ;
traffic network ;
aggregation
|
Abstract:
|
The effect of routing matrices on the dependence structure of link count time series is discussed. Given long-range dependence in the underlying OD-flows and fixed routing matrices, link count time series turn out to have even stronger dependence. Nonparametric trend estimation for link count time series illustrates the phenomenon of even stronger dependence. Randomly generated time varying routing matrices are proposed to alleviate this problem.
|
Authors who are presenting talks have a * after their name.