Assessing Measures of Order Flow Toxicity and Early Warning Signals for Market Turbulence

Following the much publicized "flash crash" in the U.S. financial markets on May 6, 2010, much work has been done in terms of developing reliable warning signals for impending market stress. However, this has met with limited success, except for one measure. The VPIN, or Volume-synchronized Probability of INformed trading, metric is introduced by Easley, Lopez de Prado and O'Hara (ELO) as a real-time indicator of order flow toxicity. They find the measure useful in predicting return volatility and conclude it, indeed, may help signal impending market turmoil. The VPIN metric involves decomposing volume into active buys and sells. We use the best-bid-offer (BBO) files from the CME Group to construct highly accurate trade classification measures for the E-mini S&P 500 futures contract. Against this benchmark, the ELO Bulk Volume Classification (BVC) scheme is inferior to a standard tick rule based on individual transactions. Moreover, when VPIN is constructed from an accurate classification, it behaves in a diametrically opposite way to BVC-VPIN. We also find the latter to have forecast power for volatility solely because it generates systematic classification errors that are correlated with trading volume and return volatility. Controlling for trading intensity and volatility, the BVC-VPIN measure has no incremental predictive power for future volatility. We conclude that VPIN is not suitable for capturing order flow toxicity or signaling ensuing market turbulence