A0556
Title: Realized variance driven by order flow imbalance
Authors: Makoto Takahashi - Osaka University (Japan) [presenting]
Abstract: The order flow imbalance (OFI), defined as the imbalance between supply and demand at the best bid and ask prices, is a main driver of price changes over a short interval and potentially a signal of private information. Based on a structural vector autoregressive model with identification through heteroskedasticity, a realized variance (RV) is decomposed into OFI-correlated and -uncorrelated components. The OFI-correlated component is constructed from returns driven by OFI innovations, that is, it represents the price impact of OFIs. The ratio of this component to the total RV is a relative informativeness of OFIs with respect to total public information and can be viewed as a measure of illiquidity. For E-mini S\&P 500 futures, the OFI-correlated component increases with the total RV but its ratio decreases when the total RV increases. Additionally, OFIs are slightly less informative when major macroeconomic announcements are released. Moreover, heterogeneous autoregression results show that the decomposition improves the in-sample forecast of volatility.
