|
Activity Number:
|
339
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Tuesday, August 8, 2006 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Statistical Computing
|
| Abstract - #306873 |
|
Title:
|
Multiresolution Outlier Detection for Long-Range Dependent Time Series
|
|
Author(s):
|
Lingsong Zhang*+ and Zhengyuan Zhu
|
|
Companies:
|
The University of North Carolina at Chapel Hill and The University of North Carolina at Chapel Hill
|
|
Address:
|
Department of Statistics, Chapel Hill, NC, 27599-3260,
|
|
Keywords:
|
outlier ; long range dependence ; time series ; multiresolution analysis
|
|
Abstract:
|
In the context of Internet traffic abnormally detection, we find that some outliers in a time series are difficult to detect in a given scale while rather obvious under a different scale. We explore the outlier effect when the underlying time series have long range dependence, and conclude that testing outliers at multiple time scales helps to reveal outlier structure in the time series. We propose a new visualization tool, the outlier map, which can help identify outliers in long range dependent time series.
|
