Paper: arXiv 2306.08809

Authors: Xiaoyue Li, John M. Mulvey

Abstract

Optimal execution of a portfolio have been a challenging problem for institutional investors. Traders face the trade-off between average trading price and uncertainty, and traditional methods suffer from the curse of dimensionality. Here, we propose a four-step numerical framework for the optimal portfolio execution problem where multiple market regimes exist, with the underlying regime switching based on a Markov process. The market impact costs are modelled with a temporary part and a permanent part, where the former affects only the current trade while the latter persists. Our approach accepts impact cost functions in generic forms. First, we calculate the approximated orthogonal portfolios based on estimated impact cost functions; second, we employ dynamic program to learn the optimal selling schedule of each approximated orthogonal portfolio; third, weights of a neural network are pre-trained with the strategy suggested by previous step; last, we train the neural network to optimize on the original trading model. In our experiment of a 10-asset liquidation example with quadratic impact costs, the proposed combined method provides promising selling strategy for both CRRA (constant relative risk aversion) and mean-variance objectives. The running time is linear in the number of risky assets in the portfolio as well as in the number of trading periods. Possible improvements in running time are discussed for potential large-scale usages.

Complexity vs Empirical Score

  • Math Complexity: 8.0/10
  • Empirical Rigor: 3.5/10
  • Quadrant: Lab Rats — high math complexity, limited empirical testing

Why this score: The paper proposes a sophisticated four-step framework combining dynamic programming, neural networks, and regime-switching models, involving significant mathematical derivations and numerical methods. However, the experimental section appears to be a limited 10-asset liquidation example with synthetic data, lacking the statistical depth, robust backtesting frameworks, and real-world data validation typically expected for empirically rigorous quant finance.

Research Flowchart

  flowchart TD
  A["Research Goal<br>Optimal Portfolio Execution<br>in Regime-Switching Market"] --> B["Data & Inputs"]
  
  subgraph B
      B1["Market Data &<br>Trading Parameters"]
      B2["Regime Parameters<br>Markov Process"]
      B3["Impact Cost Functions<br>Temporary & Permanent"]
  end

  B --> C{"Four-Step Numerical Framework"}

  subgraph C ["Computational Processes"]
      C1["Step 1: Impact Approximation<br>Calculate Orthogonal Portfolios"]
      C2["Step 2: Dynamic Programming<br>Learn Optimal Selling Schedule"]
      C3["Step 3: Neural Network<br>Pre-train with DP Strategy"]
      C4["Step 4: Final Optimization<br>Train NN on Original Model"]
  end

  C --> D["Key Findings & Outcomes"]

  subgraph D
      D1["Promising Selling Strategy<br>CRRA & Mean-Variance"]
      D2["Linear Running Time<br>O(N) in Assets & Periods"]
      D3["Scalable Framework<br>for Large-Scale Usage"]
  end

Evidence Card

What the authors claim, on what data, and who has checked it. Read from the paper's own text by a language model: every quote is verbatim, nothing was re-run. How to read this card

Proposes

The paper proposes a combined numerical method that first solves approximated orthogonal portfolios with dynamic programming and then refines the solution using a neural network for optimal multi‑asset execution in a regime‑switching market.

Source sentence
First, we calculate the approximated orthogonal portfolios based on estimated impact cost functions; second, we employ dynamic program to learn the optimal selling schedule of each approximated orthogonal portfolio; third, weights of a neural network are pre-trained with the strategy suggested by previous step; last, we train the neural network to optimize on the original trading model.

Assumes 6

  • Trades affect the market with a temporary component and a permanent component. execution
    Source sentence
    We assume trades affect the market with a temporary component and a permanent one.
  • The exact functional form of temporary and permanent impact costs is not restricted. model
    Source sentence
    However, we do not rule the exact form of either.
  • Asset returns in each period are normally distributed with parameters that depend only on the current regime. statistical
    Source sentence
    We assume the returns in each period follows a normal distribution whose parameters only depend on underlying regime.
  • Regime switching occurs only at the end of each period and follows a Markov process. model
    Source sentence
    Regime switching happens only at the end of each period, and follows a Markov process.
Show 2 more
  • The current regime is directly observable to the trader, but future regimes cannot be foreseen. market
    Source sentence
    We assume the regime is directly observable to the trader, but she cannot foresee the future regime.
  • The analysis focuses solely on the selling case of a large portfolio. execution
    Source sentence
    For simplicity, we only discuss the selling case in this paper.

Data

  • Synthetic 10‑asset liquidation example with quadratic impact costs synthetic unknown
    10 assets
    Source sentence
    The return of the risky assets under Regime 1 follows a multi-variate normal dis-tribution N([0.01, 0.005, 0.005]T , 10−5 ∗   5 1 1 1 3 1 1 1 3  ), and that under Regime 2 follows N([−0.005, 0.00, −0.01]T , 10−5 ∗   8 1 1 1 5 1 1 1 5  ). Given the parameters, we create the approximated orthogonal portfolios based on the permanent transaction cost matrices, as they are a better description of market impact

Costs

Modeled Costs are part of the reported headline results.

Both temporary and permanent market impact costs are modeled; for illustration quadratic convex functions are used.

Source sentence
The market impact costs are modelled with a temporary part and a permanent part, where the former affects only the current trade while the latter persists. For illustration reasons, we present results with a quadratic convex market impact functions, but the impact can be any function of the trading amount.

Finds 3

  • The combined method (dynamic program + neural network) outperforms the equal‑trade‑per‑period benchmark in both CRRA and mean‑variance objectives. performance
    Source sentence
    The trading schedules suggested by the approximated orthogonal portfolio approach, though suboptimal, beats the benchmark of equal trades per period in many cases, and provides an advanced starting point for further machine learning improvement. The result outperforms the benchmark, and is tested to be robust.
  • Neural‑network refinement improves expected terminal wealth from $159.919 (orthogonal‑portfolio alone) to $160.829. performance
    Source sentence
    We initilize the neural network with pre-trained weights, and train 1,000 steps with Adam Optimizer based on gradient method, resulting in a selling strategy with expected terminal wealth $160.829 and expected util-ity -0.00622. Comparing to the average wealth $159.919 and expected utility -0.00626 from approximated orthogonal portfolios alone, we notice a fair improvement by utilizing the neural networks.
  • Running time of the combined method scales linearly with the number of risky assets and trading periods. performance
    Source sentence
    The running time is linear in the number of risky assets in the portfolio as well as in the number of trading periods. The running time of this combined method is expected to be linear in number of assets as well as in number of regimes, and can be even shortened by allowing parallel computing.

Checked by others

No other paper in the archive has been found to support or contradict these findings (compared with 12 related papers). Coverage is partial, so this is neither replication nor refutation.

Code

No repository by the authors found in the paper.

Extracted 2026-11-14 · same card as JSON · A language model chose these items: check the source sentence before relying on one.