Su, Yong ChernYong ChernSuSHING-YANG HUHuang, Han ChingHan ChingHuangHsieh, Jun QueiJun QueiHsieh2019-07-232019-07-232009-01-0118104967https://scholars.lib.ntu.edu.tw/handle/123456789/414564This paper explores dynamic conditional and unconditional causality relation between intraday return and order imbalance on extraordinary events. We examine the dynamics in NASDAQ speculative top losers. In this study, we introduce a multiple hypotheses nested causality method for identifying the dynamic conditional and unconditional causality relation between intraday returns and order imbalances. The volume-stratified results suggest order imbalance be a better return predictor in small trading volume quartile. The order imbalance-based trading strategies are useful from 11:30 A.M. to 2 P.M. than in other time regimes. © Yong-Chern Su, Shing-Yang Hu, Han-Ching Huang, Jun-Quei Hsieh, 2009.enCausality | Information asymmetry | Multiple hypotheses testing method | Order imbalance | Top loserjournal article2-s2.0-78349271655https://api.elsevier.com/content/abstract/scopus_id/78349271655