Portfolio Optimization Under Uncertainty

TL;DR

This paper extends classical mean-variance portfolio theory to account for uncertainty in the estimated means and covariances of asset returns, providing a practical and intuitive method for more realistic portfolio optimization.

Contribution

It introduces a simple, intuitive approach to incorporate estimation uncertainty into mean-variance portfolio optimization, enhancing real-world applicability.

Findings

Method accounts for uncertainty in mean and covariance estimates

Results are simple, intuitive, and easily incorporated into existing optimizers

Improves portfolio robustness under estimation errors

Abstract

Classical mean-variance portfolio theory tells us how to construct a portfolio of assets which has the greatest expected return for a given level of return volatility. Utility theory then allows an investor to choose the point along this efficient frontier which optimally balances her desire for excess expected return against her reluctance to bear risk. The means and covariances of the distributions of future asset returns are assumed to be known, so the only source of uncertainty is the stochastic piece of the price evolution. In the real world, we have another source of uncertainty - we estimate but don't know with certainty the means and covariances of future asset returns. This note explains how to construct mean-variance optimal portfolios of assets whose future returns have uncertain means and covariances. The result is simple in form, intuitive, and can easily be incorporated in an optimizer.