A0602
Title: Optimal liquidation with hidden orders under self-exciting market order dynamics
Authors: Ge Zhang - National University of Singapore (Singapore) [presenting]
Ying Chen - National University of Singapore (Singapore)
Zexin Wang - Imperial College London (United Kingdom)
Chao Zhou - National University of Singapore (Singapore)
Abstract: An optimal execution strategy with both hidden and limit orders in a continuous-time framework is developed where 1) market order arrivals are either Poisson process or Hawkes process, 2) limit order submissions is subject to the exposure cost, and 3) executions of limit and hidden order both incur the immediate execution cost. Under the homogeneous Poisson process, we derive a quasi-closed-form solution containing a switching time, at which the agent changes from a pure-hidden-order phase to a mixed-orders phase until termination. Under the Hawkes process with self-exciting orders, a numerical solution is derived with feedback controls. We show there is a similar two-phase strategy, except that the switching time becomes a function of the market order intensity. The theoretical model also suggests the different impact of time pressure on order size under the two phases. In particular, the hidden order size increases with the time pressure under the pure-hidden-order phase and decreases under the mixed-orders phase, while the sum of limit and hidden orders always increases. Given transaction-level data of 100 NASDAQ stocks, real data analysis shows that the Hawkes mixture strategy outperforms the pure limit order strategy with about 70\% cost reduction and the Poisson-based mixture strategy with 27\% cost reduction.
