|
Activity Number:
|
122
- Novel Statistical Methods in the Analysis of Big Data
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Monday, July 29, 2019 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Section on Statistical Computing
|
|
Abstract #304202
|
|
|
Title:
|
Leverage Score Sampling for Multidimensional Streaming Time Series
|
|
Author(s):
|
Shuyang Bai* and Rui Xie and Ping Ma and Wenxuan Zhong and Zengyan Wang
|
|
Companies:
|
University of Georgia and University of Georgia and University of Georgia and University of Georgia and University of Georgia
|
|
Keywords:
|
Streaming Data;
Time Series;
Sampling;
Leverage Score;
Asymptotic Efficiency;
Optimal Design
|
|
Abstract:
|
Real-time modeling of multidimensional streaming time series often faces computational challenges. A useful strategy is to sample sequentially a small fraction of the data stream, based on which the online estimation of the model is carried out. In the context of the vector autoregression model, we propose a sampling method based on thresholding a leverage score. In terms of least squares estimation efficiency, the method is shown to have a performance better than alternatives such as Bernoulli sampling.
|
Authors who are presenting talks have a * after their name.
