Portfolio construction as linearly constrained separable optimization

DOI10.1007/S11081-022-09748-XarXiv2103.05455OpenAlexW3134926199MaRDI QIDQ6050367FDOQ6050367

Authors: Nicholas Moehle, Jack Gindi, Stephen Boyd, Mykel J. Kochenderfer 

Publication date: 18 September 2023

Published in: Optimization and Engineering (Search for Journal in Brave)

Abstract: Mean-variance portfolio optimization problems often involve separable nonconvex terms, including penalties on capital gains, integer share constraints, and minimum position and trade sizes. We propose a heuristic algorithm for such problems based on the alternating direction method of multipliers (ADMM). This method allows for solve times in tens to hundreds of milliseconds with around 1000 securities and 100 risk factors. We also obtain a bound on the achievable performance. Our heuristic and bound are both derived from similar results for other optimization problems with a separable objective and affine equality constraints. We discuss a concrete implementation in the case where the separable terms in the objective are piecewise quadratic, and we empirically demonstrate its effectiveness for tax-aware portfolio construction.

Full work available at URL: https://arxiv.org/abs/2103.05455

Recommendations

• Optimal portfolios using linear programming models
• Linear and mixed integer programming for portfolio optimization
• Optimal portfolio selection strategies under some constraints
• Portfolio approaches for constraint optimization problems
• A dynamic programming approach to constrained portfolios
• A unified approach to portfolio optimization with linear transaction costs
• Portfolio optimization with linear and fixed transaction costs
• Individual and cooperative portfolio optimization as linear program
• Convex optimization approaches to maximally predictable portfolio selection

zbMATH Keywords

convex optimizationportfolio optimizationproximal methodsconvex envelope

Mathematics Subject Classification ID

Convex programming (90C25) Portfolio theory (91G10)

Cites Work

• OSQP: an operator splitting solver for quadratic programs
• Title not available (Why is that?)
• Distributed optimization and statistical learning via the alternating direction method of multipliers
• Convex Analysis
• Title not available (Why is that?)
• Portfolio optimization with linear and fixed transaction costs
• Linear and mixed integer programming for portfolio optimization
• Convex hull algorithms for piecewise linear-quadratic functions in computational convex analysis
• Sparse Portfolios for High-Dimensional Financial Index Tracking
• Tax-aware portfolio construction via convex optimization

Cited In (4)

• Title not available (Why is that?)
• Linear and mixed integer programming for portfolio optimization
• Mean-variance efficient large portfolios: a simple machine learning heuristic technique based on the two-fund separation theorem
• Individual and cooperative portfolio optimization as linear program

This page was built for publication: Portfolio construction as linearly constrained separable optimization

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6050367)