The economic value of forecasting and strategy gains in volatility timing

Existing research suggests that the economic gains from volatility timing stem from covariance forecasting and portfolio construction, yet their relative importance remains unclear. This paper disentangles the contributions of these two channels by jointly evaluating three covariance models, the Dynamic Conditional Correlation Generalized Autoregressive Conditional Heteroskedasticity model (DCC), the range based Multiplicative Error Model (MEM), and the Multivariate Stochastic Volatility model (MSV), together with three portfolio construction strategies, minimum variance (MIN), maximum return (MAX), and volatility managed (VM). Our results show that the MSV model consistently outperforms DCC and MEM, highlighting the role of stochastic volatility in improving covariance estimation. From a portfolio perspective, both MAX and VM deliver higher economic value than MIN. However, there is no clear evidence that either MAX or VM consistently dominates the other, suggesting that volatility managed strategies do not systematically outperform the standard mean–variance framework.