Estimating Time-Varying Networks for High-Dimensional Time Series

• Activity Number: 476
• Type: Topic Contributed
• Date/Time: Wednesday, August 12, 2015 : 8:30 AM to 10:20 AM
• Sponsor: International Chinese Statistical Association
• Author(s): Xiaohui Chen* and Mengyu Xu and Wei Biao Wu
• Companies: University of Illinois at Urbana-Champaign and The University of Chicago and The University of Chicago
• Keywords: High-dimensional statistics ; time series analysis ; time-varying networks

Abstract

In the "big data" era, high-dimensional datasets have been increasingly seen in a broad spectrum of research fields such as in neuroscience and bioinformatics. Current development of high-dimensional data analysis in statistics and machine learning communities primarily focuses on i.i.d. observations with sub-Gaussian tails. In this talk, we shall discuss estimation of second-order time-varying networks for a general class of high-dimensional nonstationary time series data. A real example on estimating financial networks based on the S&P500 data is provided.