On high frequency dynamics between information asymmetry and volatility for securities

This paper investigates the relationship between the volatility of Volume Synchronized Probability of Informed Trading (VPIN) and future short-term volatility of stock returns. We construct a transaction-signed version of VPIN (TR-VPIN) based on tick by tick data on securities traded in the Athens Stock Exchange (ASE) during the Greek sovereign debt crisis. The results show a positive and statistically significant correlation between the volatility of TR-VPIN and future short-term volatility for securities that are exposed to asymmetric information during the period under examination. This evidence expands the existent literature which shows that the absolute order imbalance forecasts absolute returns, suggesting that TR-VPIN is a real-time informative indicator of the Probability of Informed Trading (PIN) in the high frequency domain. Further, the long-range dependence between the conditional volatilities of TR-VPIN and stock returns becomes more significant as we move towards securities which display stronger long memory. This is perfectly in line with the recent empirical evidence in microstructure literature that large past shocks of flow toxicity can lead to volatility through liquidity shortages.