Detecting and Forecasting High Volume Price Impact in the Stock Market
(2016)
Files
No attached file found for this publication.
Details
- Authors
- Abstract
- We investigate predictable patterns of high Volume Price Impact (VPI) periods using trades, orders and order book high frequency data on the Euronext 100 stocks. Our methodology is built around a fixed volume discretization of our data while VPI is captured through the Amihud ratio. We consider two forecasting models: a logistic model and an extension that is adjusted to rare events. At an average 2-minute trading volume frequency, we find that extreme VPI are mainly driven by liquidity gaps in the order book. We compute alternative VPI measures that account for time-varying volatility and periodicity in the stock market to ensure the robustness of our results. Assuming the considered sample, the models’ specifications and the methodology, our empirical findings show that our models perform pretty well in forecasting high VPI. We illustrate how such models could help market participants to reduce uncertainty around execution prices.
- Affiliations
Citations
D’Hondt, C., & Moyaert, T. (2016). Detecting and Forecasting High Volume Price Impact in the Stock Market. https://hdl.handle.net/2078.5/197360