Weighted elastic net penalized mean-variance portfolio design and computation

DOI10.1137/15M1007872zbMATH Open1330.91173arXiv1502.01658OpenAlexW1719995771MaRDI QIDQ3465255FDOQ3465255

Authors: Michael Ho, Zheng Sun, Jack Xin 

Publication date: 21 January 2016

Published in: SIAM Journal on Financial Mathematics (Search for Journal in Brave)

Abstract: It is well known that the out-of-sample performance of Markowitz's mean-variance portfolio criterion can be negatively affected by estimation errors in the mean and covariance. In this paper we address the problem by regularizing the mean-variance objective function with a weighted elastic net penalty. We show that the use of this penalty can be motivated by a robust reformulation of the mean-variance criterion that directly accounts for parameter uncertainty. With this interpretation of the weighted elastic net penalty we derive data driven techniques for calibrating the weighting parameters based on the level of uncertainty in the parameter estimates. We test our proposed technique on US stock return data and our results show that the calibrated weighted elastic net penalized portfolio outperforms both the unpenalized portfolio and uniformly weighted elastic net penalized portfolio. This paper also introduces a novel Adaptive Support Split-Bregman approach which leverages the sparse nature of $el{l}_1$ penalized portfolios to efficiently compute a solution of our proposed portfolio criterion. Numerical results show that this modification to the Split-Bregman algorithm results in significant improvements in computational speed compared with other techniques.

Full work available at URL: https://arxiv.org/abs/1502.01658

Recommendations

• Sparse weighted-norm minimum variance portfolios
• Constructing optimal sparse portfolios using regularization methods
• Penalized least squares for optimal sparse portfolio selection
• Sparse and stable Markowitz portfolios
• Solving norm constrained portfolio optimization via coordinate-wise descent algorithms

zbMATH Keywords

fast computationparameter uncertaintymean-variance portfolioadaptive support split-Bregmanpairwise elastic net regularizationstatistical weight estimationweighted elastic net regularization

Mathematics Subject Classification ID

Bootstrap, jackknife and other resampling methods (62F40) Numerical optimization and variational techniques (65K10) Statistical methods; risk measures (91G70) Robustness and adaptive procedures (parametric inference) (62F35) Portfolio theory (91G10)

Cites Work

• A Fast Iterative Shrinkage-Thresholding Algorithm for Linear Inverse Problems
• The Adaptive Lasso and Its Oracle Properties
• Numerical Optimization
• Variable Selection via Nonconcave Penalized Likelihood and its Oracle Properties
• Title not available (Why is that?)
• Bootstrap methods for standard errors, confidence intervals, and other measures of statistical accuracy
• A well-conditioned estimator for large-dimensional covariance matrices
• A generalized approach to portfolio optimization: improving performance by constraining portfolio norms
• Vast portfolio selection with gross-exposure constraints
• Common risk factors in the returns on stocks and bonds
• On the adaptive elastic net with a diverging number of parameters
• The Split Bregman Method for L1-Regularized Problems
• Estimation with quadratic loss.
• Sparse and stable Markowitz portfolios
• Estimation for Markowitz Efficient Portfolios
• Stein's Estimation Rule and Its Competitors--An Empirical Bayes Approach
• Robust Portfolio Selection Problems
• Robust Shrinkage Estimation of High-Dimensional Covariance Matrices
• Robust asset allocation
• An interior-point method for a class of saddle-point problems
• Solving norm constrained portfolio optimization via coordinate-wise descent algorithms
• Constructing optimal sparse portfolios using regularization methods
• A Multilevel Iterated-Shrinkage Approach to $l_{1}$ Penalized Least-Squares Minimization
• Regularized Tyler's Scatter Estimator: Existence, Uniqueness, and Algorithms

Cited In (21)

• \(l_1\)-regularization for multi-period portfolio selection
• Fused Lasso approach in portfolio selection
• Difference-of-convex learning: directional stationarity, optimality, and sparsity
• Dynamic portfolio optimization across hidden market regimes
• Geometric Characterization of Maximum Diversification Return Portfolio via Rao’s Quadratic Entropy
• Closed-form solutions for short-term sparse portfolio optimization
• Nonconvex multi-period mean-variance portfolio optimization
• Adaptive \(l_1\)-regularization for short-selling control in portfolio selection
• Efficient differentiable quadratic programming layers: an ADMM approach
• Loss control with rank-one covariance estimate for short-term portfolio optimization
• Multi-period portfolio selection with drawdown control
• Wavelet evolutionary network for complex-constrained portfolio rebalancing
• Short-term sparse portfolio optimization based on alternating direction method of multipliers
• Optimal portfolio selections via \(\ell_{1, 2}\)-norm regularization
• Mean-variance portfolio selection for partially observed point processes
• Dynamic portfolio optimization across hidden market regimes
• BERT-based NLP techniques for classification and severity modeling in basic warranty data study
• Degenerate Preconditioned Proximal Point Algorithms
• Sparse factor model based on trend filtering
• A Krasnoselskii-Mann proximity algorithm for Markowitz portfolios with adaptive expected return level
• Online portfolio selection with long-short term forecasting

This page was built for publication: Weighted elastic net penalized mean-variance portfolio design and computation

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3465255)