Modelling and forecasting liquidity supply using semiparametric factor dynamics

Abstract

We model the dynamics of ask and bid curves in a limit order book market using a dynamic semiparametric factor model. The shape of the curves is captured by a factor structure which is estimated nonparametrically. Corresponding factor loadings are modelled jointly with best bid and best ask quotes using a vector error correction specification. Applying the framework to four stocks traded at the Australian Stock Exchange (ASX) in 2002, we show that the suggested model captures the spatial and temporal dependencies of the limit order book. We find spillover effects between both sides of the market and provide evidence for short-term quote predictability. Relating the shape of the curves to variables reflecting the current state of the market, we show that the recent liquidity demand has the strongest impact. In an extensive forecasting analysis we show that the model is successful in forecasting the liquidity supply over various time horizons during a trading day. Moreover, it is shown that the model's forecasting power can be used to improve optimal order execution strategies.

The market impact of a limit order

“…Hence, the model is particularly useful for liquid assets where most of the market activity is concentrated at the best quote levels. In this sense, the approach complements to the dynamic model for complete order book curves introduced by Härdle, Hautsch, and Mihoci (2009)…”

Liquidity, Volatility, and Flights to Safety in the U.S. Treasury Market: Evidence from a New Class of Dynamic Order Book Models

“…Likewise, Naes and Skjeltorp (2006) regress trade size and number of transactions on volatility to document the existence of a volume-volatility relationship in the Norwegian equity market. Härdle, Hautsch, and Mihoci (2009) propose a dynamic semiparametric factor approach to modeling liquidity supply, combining non-parametric factor decomposition for the order curve's spatial structure with VAR for time variations of factor loadings. Other studies similar in their use of VAR include Danielsson and Payne (2010) and Hautsch and Huang (2009), among others.…”

Data science and digital society

“…This includes sentiment-driven stock-market reactions lifetime and liquidity of crypto-currencies, the volatility of high-frequency financial data, see , Härdle et al (2016), Härdle et al (2012).…”