Paper: arXiv 2412.19462

Authors: J. Chen, S. D. Ahipaşaoğlu, N. Zhang, Y. Yang

Abstract

We extend the classical mean-variance (MV) framework and propose a robust and sparse portfolio selection model incorporating an ellipsoidal uncertainty set to reduce the impact of estimation errors and fixed transaction costs to penalize over-diversification. In the literature, the MV model under fixed transaction costs is referred to as the sparse or cardinality-constrained MV optimization, which is a mixed integer problem and is challenging to solve when the number of assets is large. We develop an efficient semismooth Newton-based proximal difference-of-convex algorithm to solve the proposed model and prove its convergence to at least a local minimizer with a locally linear convergence rate. We explore properties of the robust and sparse portfolio both analytically and numerically. In particular, we show that the MV optimization is indeed a robust procedure as long as an investor makes the proper choice on the risk-aversion coefficient. We contribute to the literature by proving that there is a one-to-one correspondence between the risk-aversion coefficient and the level of robustness. Moreover, we characterize how the number of traded assets changes with respect to the interaction between the level of uncertainty on model parameters and the magnitude of transaction cost.

Complexity vs Empirical Score

  • Math Complexity: 8.5/10
  • Empirical Rigor: 7.0/10
  • Quadrant: Holy Grail — high math complexity, high empirical rigor

Why this score: The paper employs advanced mathematical concepts including semismooth Newton methods, difference-of-convex algorithms, and theoretical convergence proofs for an NP-hard problem, indicating high mathematical complexity. It also demonstrates strong empirical rigor through algorithm development, computational scalability tests against CPLEX, and validation using real-world datasets from Fama-French.

Research Flowchart

  flowchart TD
  A["Research Goal<br>Develop a robust, sparse<br>portfolio model to handle<br>estimation errors & costs"] --> B["Methodology<br>Extend Mean-Variance framework<br>with Ellipsoidal Uncertainty Set<br>and Fixed Transaction Costs"]

  B --> C["Data & Inputs<br>Asset returns, parameters<br>for uncertainty & cost"]

  C --> D["Computational Process<br>Semismooth Newton-based<br>Proximal DCA Algorithm<br>(Solves Mixed Integer Problem)"]

  D --> E["Key Findings & Outcomes<br>1. Analytical Link: 1-to-1<br>correspondence between risk<br>aversion & robustness<br>2. Dynamic Sparsity: Number<br>of assets adapts to uncertainty<br>& transaction cost levels<br>3. Efficient Solver: Proven<br>convergence to local minimizer"]

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 robust and sparse mean‑variance (RSMV) portfolio selection model that incorporates an ellipsoidal uncertainty set and fixed transaction costs to control estimation error and over‑diversification.

Source sentence
We extend the classical mean-variance (MV) framework and propose a robust and sparse portfolio selection model incorporating an ellipsoidal uncertainty set to reduce the impact of estimation errors and fixed transaction costs to penalize over-diversification.

Assumes 6

  • Asset returns are modeled with an ellipsoidal uncertainty set around the mean and covariance estimates. model
    Source sentence
    We extend the classical mean-variance (MV) framework and propose a robust and sparse portfolio selection model incorporating an ellipsoidal uncertainty set to reduce the impact of estimation errors and fixed transaction costs to penalize over-diversification.
  • Fixed transaction costs are incurred per traded asset and are added as a linear penalty term. execution
    Source sentence
    We extend the classical mean-variance (MV) framework and propose a robust and sparse portfolio selection model incorporating an ellipsoidal uncertainty set to reduce the impact of estimation errors and fixed transaction costs to penalize over-diversification. Therefore, the objective of RSMV model is to select a portfolio by balancing its variance, WVaR, and fixed transaction costs.
  • A parameterized covariance matrix is used to analyze how cardinality changes with the uncertainty level ε. model
    Source sentence
    Specifically, under a parameterized covariance matrix, we characterize conditions under which the cardinality of the portfolio weights might increase, decrease, or remain the same when ε increases.
  • When φ = 0 the model reduces to a robust MV optimization without transaction costs. model
    Source sentence
    Robust effects In this section, we focus on the robust MV optimization when ϕ = 0, given by RMV := min x∈C κ xT Σx + √ε √ xT Σx −¯rT x.
Show 2 more
  • The risk‑aversion coefficient κ is assumed to be positive and chosen appropriately to achieve robustness. model
    Source sentence
    In particular, we show that the MV optimization is indeed a robust procedure as long as an investor makes the proper choice on the risk-aversion coefficient. We contribute to the literature by proving that there is a one-to-one correspondence between the risk-aversion coefficient and the level of robustness.
  • The analysis of cardinality relies on conditions C1‑C3 that compare the increase Δ of uncertainty to bounds B⁻(s*) and B⁺(s*). model
    Source sentence
    The condition that s′ is smaller than s⋆is given by C1 : 0 < ∆≤min{B−(s⋆), B+(s⋆)}, the condition that s′ is equal to s⋆is given by C2 : max{0, B−(s⋆)} ≤∆≤B+(s⋆), and the condition that s′ is greater than s⋆is given by C3 : ∆≥max{0, B−(s⋆), B+(s⋆)}, where B−(s⋆) = ρ −δ −1 − 1 κσ2 min l<s⋆ l Ps⋆ i=1 ¯r[i] −s⋆Pl i=1 ¯r[i] s⋆−l −ϕs⋆l ! , and B+(s⋆) = ρ −δ −1 + 1 κσ2 min l>s⋆ l Ps⋆ i=1 ¯r[i] −s⋆Pl i=1 ¯r[i] l −s⋆ + ϕs⋆l

Data

  • FFInd12 · Fama‑French Data Library market-prices public
    01/2013‑12/2022 · 12 industry portfolios
    Source sentence
    We generate RSMV examples using monthly data from the Fama-French Data Library (FF), the Standard & Poor’s 500 stocks (SPX) and the Russell 2000 stocks (RUT) for estimating the mean and the covariance matrix. 21 Dataset Number of assets Sample Period Frequency Dataset Number of assets Sample Period Frequency FFInd12 12 01/2013-12/2022 Monthly SPX326 326 01/2013-12/2022 Daily FFInd17 17 01/2013-12/2022 Monthly RUT500
  • FFInd17 · Fama‑French Data Library market-prices public
    01/2013‑12/2022 · 17 industry portfolios
    Source sentence
    We generate RSMV examples using monthly data from the Fama-French Data Library (FF), the Standard & Poor’s 500 stocks (SPX) and the Russell 2000 stocks (RUT) for estimating the mean and the covariance matrix. 21 Dataset Number of assets Sample Period Frequency Dataset Number of assets Sample Period Frequency FFInd12 12 01/2013-12/2022 Monthly SPX326 326 01/2013-12/2022 Daily FFInd17 17 01/2013-12/2022 Monthly RUT500
  • FFInd30 · Fama‑French Data Library market-prices public
    01/2013‑12/2022 · 30 industry portfolios
    Source sentence
    We generate RSMV examples using monthly data from the Fama-French Data Library (FF), the Standard & Poor’s 500 stocks (SPX) and the Russell 2000 stocks (RUT) for estimating the mean and the covariance matrix. 21 Dataset Number of assets Sample Period Frequency Dataset Number of assets Sample Period Frequency FFInd12 12 01/2013-12/2022 Monthly SPX326 326 01/2013-12/2022 Daily FFInd17 17 01/2013-12/2022 Monthly RUT500
  • FFInd48 · Fama‑French Data Library market-prices public
    01/2013‑12/2022 · 48 industry portfolios
    Source sentence
    We generate RSMV examples using monthly data from the Fama-French Data Library (FF), the Standard & Poor’s 500 stocks (SPX) and the Russell 2000 stocks (RUT) for estimating the mean and the covariance matrix. 21 Dataset Number of assets Sample Period Frequency Dataset Number of assets Sample Period Frequency FFInd12 12 01/2013-12/2022 Monthly SPX326 326 01/2013-12/2022 Daily FFInd17 17 01/2013-12/2022 Monthly RUT500
  • SPX · Standard & Poor’s 500 market-prices public
    01/2013‑12/2022 · 326 S&P 500 constituents
    Source sentence
    We generate RSMV examples using monthly data from the Fama-French Data Library (FF), the Standard & Poor’s 500 stocks (SPX) and the Russell 2000 stocks (RUT) for estimating the mean and the covariance matrix. As a consequence, there are 326 assets in the SPX dataset and 1074 assets in the RUT dataset.
  • RUT500 · Russell 2000 market-prices public
    01/2014‑12/2018 · 500 randomly selected Russell 2000 stocks
    Source sentence
    We generate RSMV examples using monthly data from the Fama-French Data Library (FF), the Standard & Poor’s 500 stocks (SPX) and the Russell 2000 stocks (RUT) for estimating the mean and the covariance matrix. Datasets RUT500 and RUT800 are constructed by randomly selecting 500 and 800 assets from the RUT dataset, respectively.
  • RUT800 · Russell 2000 market-prices public
    01/2014‑12/2018 · 800 randomly selected Russell 2000 stocks
    Source sentence
    We generate RSMV examples using monthly data from the Fama-French Data Library (FF), the Standard & Poor’s 500 stocks (SPX) and the Russell 2000 stocks (RUT) for estimating the mean and the covariance matrix. Datasets RUT500 and RUT800 are constructed by randomly selecting 500 and 800 assets from the RUT dataset, respectively.
  • RUT1074 · Russell 2000 market-prices public
    01/2014‑12/2018 · 1074 Russell 2000 constituents
    Source sentence
    We generate RSMV examples using monthly data from the Fama-French Data Library (FF), the Standard & Poor’s 500 stocks (SPX) and the Russell 2000 stocks (RUT) for estimating the mean and the covariance matrix. As a consequence, there are 326 assets in the SPX dataset and 1074 assets in the RUT dataset.

Recognised sources: Kenneth French Data Library

Costs

Modeled Costs are part of the reported headline results.

A fixed transaction cost term ϕ·1(x) is included in the objective to penalize each non‑zero position.

Source sentence
We extend the classical mean-variance (MV) framework and propose a robust and sparse portfolio selection model incorporating an ellipsoidal uncertainty set to reduce the impact of estimation errors and fixed transaction costs to penalize over-diversification. Therefore, the objective of RSMV model is to select a portfolio by balancing its variance, WVaR, and fixed transaction costs.

Finds 5

  • The SN‑pDCA algorithm obtains solutions whose relative error to the exact CPLEX solution is less than 10%, which is less than half the error of the benchmark ℓ1‑norm portfolio. performance
    Source sentence
    Using datasets available in the Fama-French Data Library1, the relative error of the SN-pDCA solution is less than 10% compared to the CPLEX solution, which is also less than half of the relative error of the benchmark ℓ1 portfolio.
  • For 100 assets, CPLEX requires about 10 minutes while SN‑pDCA solves the problem in less than 1 second, demonstrating superior scalability. performance
    Source sentence
    With regard to the computational scalability, as the number of assets becomes 100, the computational time of CPLEX increases significantly and reaches 10 minutes. In contrast, it takes less than 1 second for the SN-pDCA to generate a suboptimal solution.
  • The set of non‑zero positions in the SN‑pDCA portfolio is a subset of those in the benchmark ℓ1 portfolio, enabling dimensionality reduction. theoretical
    Source sentence
    Moreover, the set of nonzero positions in the SN-pDCA portfolio is a subset of 1http://mba.tuck.dartmouth.edu/pages/faculty/ken.french/data library.html 4 that of the benchmark ℓ1 portfolio. Hence, the computational speed of the SN-pDCA can be further improved if we reduce the model dimension according to the actively traded assets in the benchmark ℓ1 portfolio.
  • Analytical results show a one‑to‑one correspondence between the risk‑aversion coefficient and the level of robustness, and characterize how uncertainty and transaction costs affect portfolio cardinality. theoretical
    Source sentence
    We contribute to the literature by proving that there is a one-to-one correspondence between the risk-aversion coefficient and the level of robustness. Moreover, we characterize how the number of traded assets changes with respect to the interaction between the level of uncertainty on model parameters and the magnitude of transaction cost.
  • The SN‑pDCA algorithm converges to a local minimizer of the RSMV model with a locally linear convergence rate, and under proper parameters can achieve global optimality. theoretical
    Source sentence
    We develop an efficient semismooth Newton-based proximal difference-of-convex algorithm to solve the proposed model and prove its convergence to at least a local minimizer with a locally linear convergence rate. We provide theoretical analyses to guarantee global convergence with a local linear convergence rate to the local minimizer of RSMV.

Checked by others

No other paper in the archive has been found to support or contradict these findings (compared with 30 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.

Cited References (69)

  1. Optimal and naive diversification in currency markets (2017)
  2. Robust optimization and portfolio selection: The cost of robustness (2011)
  3. Distributionally robust portfolio selection with transaction costs (2022)
  4. Portfolio Selection with Regularization (2022)
  5. An application of sparse-group lasso regularization to equity portfolio optimization and sector selection (2020)
  6. Distributionally robust mean-variance portfolio selection with Wasserstein distances (2022)
  7. A scalable algorithm for sparse portfolio selection (2022)
  8. Algorithm for cardinality-constrained quadratic optimization (2009)
and 61 more
  1. On the sensitivity of mean-variance-efficient portfolios to changes in asset means: some analytical and computational results (1991)
  2. An algorithm for portfolio optimization with transaction costs (2005)
  3. Computational study of a family of mixed-integer quadratic programming problems (1995)
  4. An exact solution approach for portfolio optimization problems under stochastic and integer constraints (2009)
  5. Keynes meets Markowitz : The trade-off between familiarity and diversification (2012)
  6. Enhancing sparsity by reweighted ℓ 1 minimization (2008)
  7. Incorporating estimation errors into portfolio selection: Robust portfolio construction (2006)
  8. Beyond Markowitz: a comprehensive wealth allocation framework for individual investors (2005)
  9. The effect of errors in means, variances, and covariances on optimal portfolio choice (2013)
  10. Portfolio selection with transaction costs (1990)
  11. A generalized approach to portfolio optimization: Improving performance by constraining portfolio norms (2009)
  12. Optimal versus naive diversification: How inefficient is the 1/N portfolio strategy? (2009)
  13. Robust portfolio optimization and management (2007)
  14. A rank minimization heuristic with application to minimum order system approximation (2001)
  15. Optimal cardinality constrained portfolio selection (2013)
  16. Portfolio selection with parameter and model uncertainty: A multi-prior approach (2007)
  17. Dynamic trading with predictable returns and transaction costs (2013)
  18. Worst-case value-at-risk and robust portfolio optimization: A conic programming approach (2003)
  19. Robust portfolio selection problems (2003)
  20. Weighted elastic net penalized mean-variance portfolio design and computation (2015)
  21. Portfolio concentration and the performance of individual investors (2008)
  22. Risk reduction in large portfolios: Why imposing the wrong constraints helps (2003)
  23. Bayes-Stein estimation for portfolio analysis (1986)
  24. Deciphering robust portfolios (2014)
  25. Portfolio optimization with linear and fixed transaction costs (2007)
  26. Portfolio Selection: Efficient Diversification of Investments (1959)
  27. Incorporating asymmetric distributional information in robust value-at-risk optimization (2008)
  28. A simple algorithm for optimal portfolio selection with fixed transaction costs (1982)
  29. The worst-case risk of a portfolio (2000)
  30. Can robust portfolio optimisation help to build better portfolios? (2007)
  31. How many stocks make a diversified portfolio? (1987)
  32. Markowitz meets Talmud: A combination of sophisticated and naive diversification strategies (2011)
  33. Robust asset allocation (2004)
  34. Improving the performance of MIQP solvers for quadratic programs with cardinality and minimum threshold constraints: A semidefinite program approach (2014)
  35. Lagrangian relaxation procedure for cardinality-constrained portfolio optimization (2008)
  36. Regularization and variable selection via the elastic net (2005)
  37. Worst-case value at risk of nonlinear portfolios (2013)
  38. Analysis of multi-stage convex relaxation for sparse regularization. (2010)
  39. A smoothing proximal gradient algorithm for nonsmooth convex regression with cardinality penalty (2020)
  40. Computing B-stationary points of nonsmooth DC programs (2017)
  41. Proximal alternating minimization and projection methods for nonconvex problems: An approach based on the Kurdyka- \L ojasiewicz inequality (2010)
  42. Convergence of descent methods for semi-algebraic and tame problems: proximal algorithms, forward–backward splitting, and regularized Gauss–Seidel methods (2013)
  43. Calculus of the exponent of Kurdyka– \L ojasiewicz inequality and its applications to linear convergence of first-order methods (2018)
  44. Variational Analysis (2009)
  45. A Newton-CG augmented Lagrangian method for semidefinite programming (2010)
  46. Matrix Computations, 3rd ed. (1996)
  47. Convex Analysis (1996)
  48. On the convergence of the proximal algorithm for nonsmooth functions involving analytic features (2009)
  49. An efficient Hessian based algorithm for solving large-scale sparse group Lasso problems (2020)
  50. A highly efficient semismooth Newton augmented Lagrangian method for solving Lasso problems (2018)
  51. First-order Methods in Optimization (2017)
  52. Optimization and nonsmooth analysis (1990)
  53. Variable selection via nonconcave penalized likelihood and its oracle properties (2001)
  54. Nearly unbiased variable selection under minimax concave penalty (2010)
  55. High-Dimensional Distributionally Robust Mean-Variance Efficient Portfolio Selection (2023)
  56. Difference-of-convex learning: directional stationarity, optimality, and sparsity (2017)
  57. DC programming and DCA : thirty years of developments (2018)
  58. Convex analysis approach to DC programming: theory, algorithms and applications (1997)
  59. Regression shrinkage and selection via the lasso (1996)
  60. Sparse and stable Markowitz portfolios (2009)
  61. Constructing optimal sparse portfolios using regularization methods (2015)

0 of 69 cited works are in the corpus; resolved via arXiv source bundles, Semantic Scholar and OpenAlex. Corpus-wide: 2596 linked papers.