Paper: arXiv 2305.16152

Authors: Areski Cousin, Jérôme Lelong, Tom Picard

Abstract

This paper studies the multi-period mean-variance portfolio allocation problem with transaction costs. Many methods have been proposed these last years to challenge the famous uni-period Markowitz strategy.But these methods cannot integrate transaction costs or become computationally heavy and hardly applicable. In this paper, we try to tackle this allocation problem by proposing an innovative approach which relies on representing the set of admissible portfolios by a finite dimensional Wiener chaos expansion. This numerical method is able to find an optimal strategy for the allocation problem subject to transaction costs. To complete the study, the link between optimal portfolios submitted to transaction costs and the underlying risk aversion is investigated. Then a competitive and compliant benchmark based on the sequential uni-period Markowitz strategy is built to highlight the efficiency of our approach.

Complexity vs Empirical Score

  • Math Complexity: 9.0/10
  • Empirical Rigor: 3.0/10
  • Quadrant: Lab Rats — high math complexity, limited empirical testing

Why this score: The paper employs advanced stochastic analysis and Wiener chaos expansions, involving heavy mathematical derivations and theoretical frameworks, while presenting no empirical data, code, or backtesting results, focusing solely on theoretical propositions and computational method design.

Research Flowchart

  flowchart TD
  A["Research Goal:<br>Multi-period Mean-Variance<br>Portfolio Optimization with<br>Transaction Costs"] --> B{"Key Methodology"};
  B --> C["Wiener Chaos Expansion<br>for Admissible Portfolios"];
  B --> D["Sequential Single-Period<br>Markowitz Benchmark"];
  
  C --> E["Computational Process:<br>Finite Dimensional Approximation<br>Optimal Strategy Derivation"];
  D --> E;
  
  E --> F["Data/Inputs:<br>Asset Return Scenarios<br>Risk Aversion Parameters<br>Transaction Cost Structures"];
  
  F --> G{"Key Findings/Outcomes"};
  G --> H["Efficient Optimal Strategy<br>Integrated Transaction Costs"];
  G --> I["Risk Aversion Impact<br>Analysis on Portfolio Allocation"];
  G --> J["Competitive Benchmark<br>Validating Approach Efficiency"];

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 an innovative numerical method based on a finite‑dimensional Wiener chaos expansion to obtain optimal multi‑period mean‑variance portfolios under transaction costs.

Source sentence
we try to tackle this allocation problem by proposing an innovative approach which relies on representing the set of admissible portfolios by a finite dimensional Wiener chaos expansion.

Assumes 4

  • The financial market is complete. market
    Source sentence
    We assume that the financial market they represent is complete.
  • The risk‑free rate is deterministic. model
    Source sentence
    The risk free rate (rn)n≤N is assumed to be deterministic and ∀n
  • Transaction costs are proportional to the traded cash volume per asset with constant rate ν. execution
    Source sentence
    He pays the transaction costs, proportional to the trade cash volume per asset and equal to
  • The first position is assumed to be free of charge when costs are present. execution
    Source sentence
    In presence of costs, we assume that the first position is free of charge.

Data

  • Synthetic multi‑asset model synthetic unknown
    3 risky assets
    Source sentence
    ... µ 0.06 0.02 0.14 ... ˆσ 0.1 0.06 0.2 ... Table 2: Correlation matrix ρ of risky assets

Costs

Modeled Costs are part of the reported headline results.

Proportional cost ν per unit of traded cash volume for each asset, applied in the self‑financing condition and in the optimization.

Source sentence
He pays the transaction costs, proportional to the trade cash volume per asset and equal to

Finds 3

  • The Sharpe ratio of a locally optimal portfolio with transaction costs does not depend on its risk aversion. theoretical
    Source sentence
    we have proved that the Sharpe ratio of a locally optimal portfolio with transaction costs does not depend on its risk aversion.
  • All optimal portfolios under transaction costs share the same Sharpe ratio. theoretical
    Source sentence
    we have established that all optimal portfolios have the same Sharpe ratio.
  • The multi‑period Wiener‑chaos method outperforms the sequential uni‑period Markowitz benchmark. performance
    Source sentence
    As expected, the multi-period model is more intricate, but outperforms the uni-period models.

Checked by others

No other paper in the archive has been found to support or contradict these findings (compared with 18 related papers). Coverage is partial, so this is neither replication nor refutation.

Cited by 1 paper in the archive (1 by independent authors)

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.