RELATION BETWEEN ORDER IMBALANCE AND RETURNS ON NYSE

Abstract

We study the relation between limit order imbalance and intraday return, using limit order book for New York Stock Exchange (NYSE). Our study is motivated by the desire to better understand the interaction between stock price movement and trading activity, in particular order imbalance. We define limit order imbalance as weighted total number of orders on the bid side minus weighted total number of sell orders on the limit order book.We investigate the dynamic relation between order imbalance and return using vectorautoregressive model. We find limit order imbalances are positively autocorrelated, consistent with order splitting and herding behavior. The persistent order imbalances have positive impact on price movements, and this impact is even greater for large stocks. Our result also indicate that return have negative impact on order imbalance, suggesting investors place more sell orders following a price increase. Lastly, we confirm the bi-directional causality between order imbalance and return, using the Granger Causality test