Integrating prediction in mean-variance portfolio optimization

2021/02/18 by Butler, Andrew, Kwon, Roy H. · 4 citations
#FOS: Economics and business #Portfolio Management (q-fin.PM)

paper · doi:10.48550/arxiv.2102.09287

Abstract

Prediction models are traditionally optimized independently from their use in the asset allocation decision-making process. We address this shortcoming and present a framework for integrating regression prediction models in a mean-variance optimization (MVO) setting. Closed-form analytical solutions are provided for the unconstrained and equality constrained MVO case. For the general inequality constrained case, we make use of recent advances in neural-network architecture for efficient optimization of batch quadratic-programs. To our knowledge, this is the first rigorous study of integrating prediction in a mean-variance portfolio optimization setting. We present several historical simulations using both synthetic and global futures data to demonstrate the benefits of the integrated approach.

Cited by

• Estimating Covariance for Global Minimum Variance Portfolio: A Decision-Focused Learning Approach
• End-to-End Large Portfolio Optimization for Variance Minimization with Neural Networks through Covariance Cleaning
• Decision-Induced Ranking Explains Prediction Inflation and Excessive Turnover in SPO-Based Portfolio Optimization
• Enhancing Portfolio Optimization with Deep Learning Insights

Related

• Data-driven integration of norm-penalized mean-variance portfolios
• Return Prediction for Mean-Variance Portfolio Selection: How Decision-Focused Learning Shapes Forecasting Models
• Investment Portfolio Optimization Based on Modern Portfolio Theory and Deep Learning Models
• Integrated Prediction and Multi-period Portfolio Optimization
• A Universal End-to-End Approach to Portfolio Optimization via Deep Learning