Abstract
Market liquidity is a latent and dynamic variable. We propose a dynamical linear price impact model at high-frequency in which the price impact coefficient is a product of a daily, a diurnal, and an auto-regressive stochastic intraday component. We estimate the model using a Kalman filter on order book data for stocks traded on the NASDAQ in 2016. We show that our price changes estimates conditional on order flow imbalance explain, on average, $82\%$ of real price changes variance. Evidence is also provided on the fact that the conditioning on filtered information improves the estimate of the \text{LOB} liquidity with respect to the one obtained from a static estimation of the price impact. In addition, an out-of-sample analysis shows that our model provides a superior out-of-sample forecast of price impact with respect to historical estimates.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have