VPIN and extreme market conditions: evidence from Taiwan index futures market

• Main Authors: Kang-Syuan Peng, 彭康軒
• Other Authors: JUN-MAO CHIU
• Format: Others
• Language: en_US
• Online Access: http://ndltd.ncl.edu.tw/handle/37qsfc

碩士 === 元智大學 === 財務金融暨會計碩士班(財務金融學程) === 103 === This research use the new procedure presented by Easley et al. (2012a) to estimate flow toxicity, which named volume-synchronized probability of informed trading, or the VPIN flow-toxicity metric. The VPIN metric is based on volume imbalance and trade intensity which can more accurate to predict the volatility in the high frequency trading (HFT) environment. Because in HFT, trade volume convey more information than clock-time. We use the Taiwan index futures’ (TX) data from 2007/01/01 to 2008/12/31 to estimate the VPIN metric to see whether the VPIN metric can predict TX’s volatility. Then, we define some extreme market conditions, including macroeconomics announcements, extreme volatility, and extreme VIX. We find that if it is the one bucket before extreme volatility and extreme VIX, it VPIN are more higher than the others buckets. But this results not happened in the pre-macroeconomics announcements. We also find the evidence the VPIN metric can forecast TX’s volatility, especially in the financial crisis period this phenomenon is more significant, but we find a little evidence that the VPIN metric can forecast the VIX.