Paper: arXiv 2602.14439

Authors: Ruike Wu, Yonghe Lu, Yanrong Yang

Abstract

This paper investigates the impact of environmental, social, and governance (ESG) constraint on a regularized mean-variance (MV) portfolio optimization problem in a large-dimensional setting, in which a positive definite regularization matrix is imposed on the sample covariance matrix. We first derive the asymptotic results for the out-of-sample (OOS) Sharpe ratio (SR) of the proposed portfolio, which help quantify the impact of imposing an ESG-level constraint as well as the effect of estimation error arising from the sample mean estimation of the assets’ ESG score. Furthermore, to study the influence of the choices of the regularization matrix, we develop an estimator for the OOS Sharpe ratio. The corresponding asymptotic properties of the Sharpe ratio estimator are established based on random matrix theory. Simulation results show that the proposed estimators perform close to the corresponding oracle level. Moreover, we numerically investigate the impact of various forms of regularization matrices on the OOS SR, which provides useful guidance for practical implementation. Finally, based on OOS SR estimator, we propose an adaptive regularized portfolio which uses the best regularization matrix yielding the highest estimated SR (among a set of candidates) at each decision node. Empirical evidence based on the S&P 500 index demonstrates that the proposed adaptive ESG-constrained portfolio achieves a high OOS SR while satisfying the required ESG level, offering a practically effective approach for sustainable investment.

Complexity vs Empirical Score

  • Math Complexity: 8.0/10
  • Empirical Rigor: 7.5/10
  • Quadrant: Holy Grail — high math complexity, high empirical rigor

Why this score: The paper provides a strong theoretical framework using random matrix theory to address estimation error in large-dimensional ESG-constrained portfolios. It complements rigorous asymptotic proofs with comprehensive Monte Carlo simulations and an empirical backtest on S&P 500 components.

Research Flowchart

  flowchart TD
    A[Goal: Impact of ESG Constraints on Regularized MV Portfolio in High Dimensions] --> B[Input: S&P 500 Data, Asset ESG Scores & Covariance Matrix]
    B --> C[Method: Derive Asymptotic OOS Sharpe Ratio via Random Matrix Theory]
    C --> D[Develop OOS Sharpe Ratio Estimator for Regularization Matrices]
    D --> E[Simulations & Empirical Evaluation: Select Best Matrix per Decision Node]
    E --> F[Outcome: Adaptive ESG Portfolio Maximizes OOS SR While Meeting ESG Target]

Cited References (179)

  1. Risk reduction in large portfolios: Why imposing the wrong constraints helps (2003)
  2. Recent advances in shrinkage-based high-dimensional inference (2022)
  3. Optimal shrinkage estimator for high-dimensional mean vector (2019)
  4. Sustainability-valued discrete option pricing in complete markets (2025)
  5. Cost-aware Portfolios in a Large Universe of Assets — → 2412.11575
  6. Large-scale portfolio allocation under transaction costs and model uncertainty (2019)
  7. In-sample and out-of-sample Sharpe ratios of multi-factor asset pricing models (2024)
  8. Double Descent in Portfolio Optimization: Dance between Theoretical Sharpe Ratio and Estimation Accuracy — → 2411.18830
and 171 more
  1. Cross-Sectional Learning and Inference for the Stochastic Discount Factor (2025)
  2. Markowitz revisited: Social portfolio engineering (2017)
  3. Optimal ESG portfolios: an example for the Dow Jones Index (2022)
  4. Environmental, Social and Governance-Valued Portfolio Optimization and Dynamic Asset Pricing (2025)
  5. Making distributionally robust portfolios feasible in high dimension (2025)
  6. A Shrinkage Covariance Matrix Aligned with Sentiment (2025)
  7. Machine learning and factor-based portfolio optimization (2021)
  8. A Dynamic Semiparametric Characteristics-based Model for Portfolio Selection (2024)
  9. Semiparametric conditional factor models: Estimation and inference (2023)
  10. Dynamic peer groups of arbitrage characteristics (2024)
  11. Instrumented principal component analysis (2017)
  12. Optimal portfolio diversification via independent component analysis (2022)
  13. Autoencoder asset pricing models (2021)
  14. Semiparametric estimation of a characteristic-based factor model of common stock returns (2007)
  15. Efficient semiparametric estimation of the Fama–French model and extensions (2012)
  16. Empirical tests of asset pricing models with individual assets: Resolving the errors-in-variables bias in risk premium estimation (2019)
  17. Comparing cross-section and time-series factor models (2020)
  18. Extra-market components of covariance in security returns (1974)
  19. Robust Large Portfolio Optimization with Heteroscedastic and Heavy-Tailed Returns (2025)
  20. Optimal versus naive diversification: H ow inefficient is the $ 1/N $ portfolio strategy? (2009)
  21. A generalized approach to portfolio optimization: Improving performance by constraining portfolio norms (2009)
  22. Vast portfolio selection with gross-exposure constraints (2012)
  23. Portfolio Selection (1952)
  24. Enhancement of the applicability of Markowitz’s portfolio optimization by utilizing random matrix theory (2009)
  25. Putting Markowitz theory to work (1981)
  26. The M arkowitz optimization enigma: Is ‘optimized’optimal? (1989)
  27. On the Sensitivity of Mean-Variance-Efficient Portfolios to Changes in Asset Means: Some Analytical and Computational Results (1991)
  28. International portfolio diversification with estimation risk (1985)
  29. CLT for linear spectral statistics of large-dimensional sample covariance matrices (2004)
  30. Spectral analysis of large dimensional random matrices (2010)
  31. Strong Convergence of the Empirical Distribution of Eigenvalues of Large Dimensional Random Matrices (1995)
  32. The risk of expected utility under parameter uncertainty (2024)
  33. Multivariate shrinkage for optimal portfolio weights (2007)
  34. Dominating estimators for minimum-variance portfolios (2010)
  35. Why Naive Diversification Is Not So Naive, and How to Beat It? (2024)
  36. Tri-criterion modeling for constructing more-sustainable mutual funds (2015)
  37. Estimation of the global minimum variance portfolio in high dimensions (2018)
  38. A well-conditioned estimator for large-dimensional covariance matrices (2004)
  39. Predicting returns with text data (2019)
  40. Optimal shrinkage-based portfolio selection in high dimensions (2022)
  41. Shocks-adaptive Robust Minimum Variance Portfolio for a Large Universe of Assets — → 2410.01826
  42. Markowitz meets Talmud: A combination of sophisticated and naive diversification strategies (2011)
  43. Optimal portfolio choice with estimation risk: No risk-free asset case (2022)
  44. Testing Independence Among a Large Number of High-Dimensional Random Vectors (2014)
  45. High-dimensionality effects in the Markowitz problem and other quadratic programs with linear constraints: Risk underestimation (2010)
  46. High dimensional minimum variance portfolio estimation under statistical factor models (2021)
  47. Approaching mean-variance efficiency for large portfolios (2019)
  48. Large covariance estimation by thresholding principal orthogonal complements (2013)
  49. Optimal Portfolio Using Factor Graphical Lasso (2020) — link
  50. Sparsistency and rates of convergence in large covariance matrix estimation (2009)
  51. Text-based network industries and endogenous product differentiation (2016)
  52. On time-varying factor models: Estimation and testing (2017)
  53. A constrained ℓ1 minimization approach to sparse precision matrix estimation (2011)
  54. Adaptive thresholding for sparse covariance matrix estimation (2011)
  55. Covariance regularization by thresholding (2008)
  56. High dimensional covariance matrix estimation using a factor model (2008)
  57. Nonlinear shrinkage of the covariance matrix for portfolio selection: M arkowitz meets G oldilocks (2017)
  58. A nodewise regression approach to estimating large portfolios (2021)
  59. Sharpe Ratio in High Dimensions: Cases of Maximum Out of Sample, Constrained Maximum, and Optimal Portfolio Choice (2020) — link
  60. The arbitrage theory of capital asset pricing (1976)
  61. The Cross-Section of Expected Stock Returns (1992)
  62. Common risk factors in the returns on stocks and bonds (1993)
  63. A five-factor asset pricing model (2015)
  64. Taming the factor zoo: A test of new factors (2020)
  65. Lucky factors (2021)
  66. Nonparametric specification testing for continuous-time models with applications to term structure of interest rates (2005)
  67. Sparse estimation of a covariance matrix (2011)
  68. Sparse covariance matrix estimation by DCA-based algorithms (2017)
  69. Testing for structural changes in factor models via a nonparametric regression (2020)
  70. Large dynamic covariance matrices (2019)
  71. The adaptive lasso and its oracle properties (2006)
  72. Determining the number of factors in approximate factor models (2002)
  73. Improved estimation of the covariance matrix of stock returns with an application to portfolio selection (2003)
  74. Large system of seemingly unrelated regressions: A penalized quasi-maximum likelihood estimation perspective (2020)
  75. Nonparametric estimation of large covariance matrices with conditional sparsity (2021)
  76. Statistical Foundations of Data Science (2020)
  77. Portfolio Selection: E fficient Diversification of Investment. Cowles Foundation for Research in Economics, Monograph 16 (1959)
  78. Hypothesis testing on linear structures of high-dimensional covariance matrix (2019)
  79. Homogeneity tests of covariance matrices with high-dimensional longitudinal data (2019)
  80. Factor modeling for high-dimensional time series: Inference for the number of factors (2012)
  81. Dynamic Conditional Correlation: A Simple Class of Multivariate Generalized Autoregressive Conditional Heteroskedasticity Models (2002)
  82. Testing Independence Among a Large Number of High-Dimensional Random Vectors (2014)
  83. High dimensional correlation matrices: the central limit theorem and its applications (2017)
  84. Structural deep learning in conditional asset pricing (2022)
  85. Bias in error estimation when using cross-validation for model selection (2006)
  86. State-Varying Factor Models of Large Dimensions (2020)
  87. High-dimensional covariance estimation by minimizing ℓ1-penalized log-determinant divergence (2011)
  88. Arbitrage, Factor Structure, and Mean-Variance Analysis on Large Asset Markets (1983)
  89. Time-varying risk premium in large cross-sectional equity data sets (2016)
  90. Asset Pricing with Omitted Factors (2021)
  91. High-dimensional minimum variance portfolio estimation based on high-frequency data (2020)
  92. High-dimensionality effects in the markowitz problem and other quadratic programs with linear constraints: risk underestimation (2010)
  93. Using principal component analysis to estimate a high dimensional factor model with high-frequency data (2017)
  94. Correlation risk and optimal portfolio choice (2010)
  95. Structural Change in Investment and Consumption–A Unified Analysis (2021)
  96. Skill-Biased Structural Change (2021)
  97. An Analytic Derivation of the Efficient Portfolio Frontier (1972)
  98. On estimating the expected return on the market: An exploratory investigation (1980)
  99. A new semiparametric estimation approach for large dynamic covariance matrices with multiple conditioning variables (2019)
  100. MEAN–VARIANCE PORTFOLIO OPTIMIZATION WITH STATE-DEPENDENT RISK AVERSION (2012)
  101. Factor Models for Portfolio Selection in Large Dimensions: The Good, the Better and the Ugly (2021)
  102. DISTRIBUTED ESTIMATION OF PRINCIPAL EIGENSPACES (2019)
  103. Optimal estimation of functionals of high-dimensional mean and covariance matrix (2019) — link
  104. Principal Components and Orthogonal Regression Based on Robust Scales (2005)
  105. Robustifying Markowitz — → 2212.13996
  106. Panel data models with interactive fixed effects (2009)
  107. Linear regression for panel with unknown number of factors as interactive fixed effects (2015)
  108. Estimating cross-section common stochastic trends in nonstationary panel data (2004)
  109. Panel cointegration with global stochastic trends (2009)
  110. Efficient Estimation of Nonstationary Factor Models (2017)
  111. A P anic Attack on Unit Roots and Cointegration (2004)
  112. Panel Data Models with Interactive Fixed Effects (2009)
  113. Varying-coefficient panel data models with nonstationarity and partially observed factor structure (2021)
  114. The effect of estimation risk on optimal portfolio choice (1976)
  115. Portfolio selection with parameter and model uncertainty: A multi-prior approach (2007)
  116. Integrating Factor Models (2022)
  117. Realized semibetas: Disentangling “good” and “bad” downside risks (2022)
  118. Dynamic Trading with Predictable Returns and Transaction Costs (2013)
  119. Portfolio Selection with Robust Estimation (2009)
  120. Time-varying minimum variance portfolio (2024)
  121. A unified model for regularized and robust portfolio optimization (2019)
  122. Inferential theory for factor models of large dimensions (2003)
  123. Challenging the empirical mean and empirical variance: A deviation study (2012)
  124. Robust covariance estimation for approximate factor models (2019)
  125. Factor Models, Machine Learning, and Asset Pricing (2022)
  126. Large covariance estimation through elliptical factor models (2018)
  127. Factor Momentum and the Momentum Factor (2022)
  128. On Persistence in Mutual Fund Performance (1997)
  129. Sparse and stable M arkowitz portfolios (2009)
  130. Weighted elastic net penalized mean-variance portfolio design and computation (2015)
  131. Distributionally robust mean-variance portfolio selection with W asserstein distances (2022)
  132. Constructing optimal sparse portfolios using regularization methods (2015)
  133. High-Dimensional Distributionally Robust Mean-Variance Efficient Portfolio Selection (2023)
  134. A synthetic regression model for large portfolio allocation (2022)
  135. A robust perspective on transaction costs in portfolio optimization (2018)
  136. The worst-case risk of a portfolio (2000)
  137. Robust portfolio selection problems (2003)
  138. Ambiguity in portfolio selection (2007)
  139. Data-driven distributionally robust optimization using the W asserstein metric: P erformance guarantees and tractable reformulations (2018)
  140. Distributionally robust optimization under moment uncertainty with application to data-driven problems (2010)
  141. Estimation of W asserstein distances in the spiked transport model (2022)
  142. Heteroskedasticity and autocorrelation consistent covariance matrix estimation (1991)
  143. Powerful tests for structural changes in volatility (2013)
  144. Consistent covariance matrix estimation for linear processes (2002)
  145. Consistency of kernel estimators of heteroscedastic and autocorrelated covariance matrices (2000)
  146. Principal components estimation and identification of static factors (2013)
  147. An improved heteroskedasticity and autocorrelation consistent covariance matrix estimator (1992)
  148. Integrative factor regression and its inference for multimodal data analysis (2022)
  149. Finite-sample guarantees for W asserstein distributionally robust optimization: B reaking the curse of dimensionality (2023)
  150. On sharpness of T chebycheff-type inequalities (1962)
  151. Robust W asserstein profile inference and applications to machine learning (2019)
  152. Optimal portfolio choice with parameter uncertainty (2007)
  153. CLT for linear spectral statistics of large-dimensional sample covariance matrices (2004)
  154. Characteristics are covariances: A unified model of risk and return (2019)
  155. Performance Attribution for Portfolio Constraints (2025)
  156. Impact investing (2021)
  157. Balancing returns and responsibility: Evidence from shrinkage-based portfolios (2023)
  158. Quantifying the Returns of ESG Investing: An Empirical Analysis with Six ESG Metrics. (2024)
  159. Quantifying the impact of impact investing (2024)
  160. ESG: A new dimension in portfolio allocation (2023)
  161. Estimation of Out-of-Sample Sharpe Ratio for High Dimensional Portfolio Optimization (2025)
  162. Projected principal component analysis in factor models (2016)
  163. Arbitrage portfolios (2021)
  164. Responsible investing: The ESG-efficient frontier (2021)
  165. Does ESG impact really enhance portfolio profitability? (2022)
  166. Eigenvectors of some large sample covariance matrix ensembles (2011)
  167. Limit of the smallest eigenvalue of a large dimensional sample covariance matrix (1993)
  168. Performance analysis and optimal selection of large minimum variance portfolios under estimation risk (2012)
  169. High-dimensionality effects in the Markowitz problem and other quadratic programs with linear constraints: Risk underestimation (2010)
  170. Spectrally-corrected estimation for high-dimensional Markowitz mean-variance optimization (2022)
  171. Two is better than one: Regularized shrinkage of large minimum variance portfolios (2024)

4 of 179 cited works are in the corpus; resolved via arXiv source bundles, Semantic Scholar and OpenAlex. Corpus-wide: 2518 linked papers.