A0736
Title: Online GMM and GEL for time series
Authors: Man Fung Leung - University of Illinois Urbana-Champaign (United States) [presenting]
Kin Wai Chan - The Chinese University of Hong Kong (Hong Kong)
Xiaofeng Shao - Washington University in St Louis (United States)
Abstract: Time series data are inherently serially dependent and sequential. Motivated by the need to analyze large-scale streaming time series data, online versions of the classical GMM and GEL are presented. The methodological development covers point estimation, confidence region construction, over-identifying restrictions testing, and anomaly detection. Connections to and differences from existing methods are also discussed. Finally, some encouraging finite sample results are shown in numerical simulations and data illustrations.
