Home > bal > AABFJ > Vol. 7 (2013) > Iss. 2
Abstract
This study involves a detailed discussion on the estimation of intraday time-varying volume synchronised probability of informed trading (VPIN), a proxy for levels of informed trading and flow toxicity, followed by intraday analysis on its impact of the behaviour of intraday trading in a limit order book (LOB) market. The variation of VPIN used is closely based on the original from Easley, Lopez de Prado and O’Hara (2010), using trade volume imbalance information. This study shows that different capitalisation stocks exhibit different VPIN characteristics. Previous studies on other variations of PIN have looked at its determination on price movements, and whether a lead-lag relationship exists. This study examines if VPIN has an effect on several of intraday trading factors in the Australian market, being a LOB. In particular, it documents if Granger causality exists between (1) VPIN and quote imbalance, 20(2) VPIN and intraday price volatility and (3) VPIN and intraday trade frequency or similarly in an inverse manner, duration.
For this analysis the Hsiao-Kang methodology for Granger testing has been followed and the posterior odds ratio test used to measure the strength of Granger causality in equity markets as suggested recently by Atukeren (2005). Feedback causality between VPIN and all three intraday factors is apparent.
