Paper: arXiv 2603.28898

Authors: Thomas P. McAuliffe, Samuel Liew, Yuchao Li, Andrey Ushenin, Chihang Wang, Alexandros Tasos, Jack Pearce, Dimitris Tasoulis, Dimitri P. Bertsekas, Theodoros Tsagaris

Abstract

We address the problem of executing large client orders in continuous double-auction markets under time and liquidity constraints. We propose a model predictive control (MPC) framework that balances three competing objectives: order completion, market impact, and opportunity cost. Our algorithm is guided by a trading schedule (such as time-weighted average price or volume-weighted average price) but allows for deviations to reduce the expected execution cost, with due regard to risk. Our MPC algorithm executes the order progressively, and at each decision step it solves a fast quadratic program that trades off expected transaction cost against schedule deviation, while incorporating a residual cost term derived from a simple base policy. Approximate schedule adherence is maintained through explicit bounds, while variance constraints on deviation provide direct risk control. The resulting system is modular, data-driven, and suitable for deployment in production trading infrastructure. Using six months of NASDAQ ’level 3’ data and simulated orders, we show that our MPC approach reduces schedule shortfall by approximately 40-50% relative to spread-crossing benchmarks and achieves significant reductions in slippage. Moreover, augmenting the base policy with predictive price information further enhances performance, highlighting the framework’s flexibility for integration with forecasting components.

Complexity vs Empirical Score

  • Math Complexity: 6.5/10
  • Empirical Rigor: 7.5/10
  • Quadrant: Holy Grail — high math complexity, high empirical rigor

Why this score: The paper presents a mathematically sound Model Predictive Control framework for trade execution, formulated as a quadratic program. It demonstrates strong empirical rigor by validating the approach on six months of high-granularity NASDAQ Level 3 data with substantial reductions in slippage and schedule shortfall.

Research Flowchart

  flowchart TD
    A["Research Goal:<br>Optimize Large Order Execution via MPC"] --> B["Inputs:<br>6 Months NASDAQ L3 Data & Target Schedules"]
    B --> C["Optimization Step:<br>Solve Fast Quadratic Program at Each Step"]
    C --> D["Constraints & Balances:<br>Market Impact vs. Schedule Bounds vs. Risk"]
    D --> E["Enhancement:<br>Integrate Predictive Price Information"]
    E --> F["Outcomes:<br>40-50% Shortfall Reduction & Lower Slippage"]

Cited References (30)

  1. The Total Cost of Transactions on the NYSE (1988)
  2. Measuring Historical Volatility: Close-to-Close, Exponentially Weighted, Parkinson, Garman-Klass, Rogers-Satchell and Yang-Zhang Volatility (2012)
  3. Volume Weighted Average Price Optimal Execution (2015) — link
  4. A Course in Reinforcement Learning (2025)
  5. The Price Impact of Order Book Events (2014) — link
  6. Machine Learning in FX — link
  7. Morgan Stanley’s U.S. Cash Equity Order Handling & Routing Practices: Frequently Asked Questions — link
  8. FX Algorithmic Execution — link
and 22 more
  1. Optimal Execution with Limit and Market Orders (2015) — link
  2. Optimal Execution with Nonlinear Impact Functions and Trading-Enhanced Risk (2003) — link
  3. Optimal Execution of Portfolio Transactions (2000)
  4. Optimal Control of Execution Costs (1998) — link
  5. A Reinforcement Learning Approach to Optimal Execution (2022) — link
  6. Reinforcement Learning for Optimized Trade Execution (2006) — link
  7. Hierarchical Deep Reinforcement Learning for VWAP Strategy Optimization — → 2212.14670
  8. A Reinforcement Learning Extension to the Almgren–Chriss Framework for Optimal Trade Execution (2014)
  9. Double Q -learning (2010)
  10. Feudal Reinforcement Learning (1993)
  11. Between MDPs and Semi- MDPs : A Framework for Temporal Abstraction in Reinforcement Learning (1999) — link
  12. Model predictive control and reinforcement learning: A unified framework based on dynamic programming (2024)
  13. Model Predictive Control and Reinforcement Learning: A Unified Framework Based on Dynamic Programming (2024)
  14. Reinforcement Learning and Optimal Control (2019)
  15. Rollout, Policy Iteration, and Distributed Reinforcement Learning (2020)
  16. Mastering the Game of Go with Deep Neural Networks and Tree Search (2016) — link
  17. A General Reinforcement Learning Algorithm that Masters Chess, Shogi, and Go through Self-Play (2018) — link
  18. TD-Gammon, a Self-Teaching Backgammon Program, Achieves Master-Level Play (1994) — link
  19. Clarabel: An interior-point solver for conic programs with quadratic objectives (2024) — link
  20. A Deep Learning Approach to Estimating Fill Probabilities in a Limit Order Book (2022) — link
  21. Stock Trading via Feedback Control: Stochastic Model Predictive or Genetic? (2017) — link
  22. Optimal trading: a model predictive control approach (2021) — link

1 of 30 cited works are in the corpus; resolved via arXiv source bundles, Semantic Scholar and OpenAlex. Corpus-wide: 2518 linked papers.