Hedging market risk and uncertainty via a robust portfolio approach

Abstract

Shorting for hedging exposes to risk when the market dynamics is uncertain. Managing uncertainty and risk exposure is key in portfolio management practice. This paper develops a robust framework for dynamic minimum-variance hedging that explicitly accounts for forecast uncertainty in volatility and covariance estimation to achieve empirical stability and reduced turnover, further improving other standard performance metrics. The approach combines high-frequency realized variance and covariance measures, autoregressive models for multi-step volatility forecasting, and a box-uncertainty robust optimization scheme. We derive a closed-form solution for the robust hedge ratio, which adjusts the standard minimum-variance hedge by incorporating variance and covariance forecast uncertainty. Using a diversified sample of equity, bond, and commodity ETFs over 2016-2024, we show that robust hedge ratios are more stable and entail lower turnover than standard dynamic hedges. While overall variance reduction is comparable, the robust approach improves downside protection and risk-adjusted performance, particularly when transaction costs are considered. Bootstrap evidence supports the statistical significance of these gains.