Forward Variance Factor Predicts Stock Returns
Log in to collectAcademic paper
International Stock Return Predictability: The Role of U.S. Volatility Risk
Yizhe Deng; Fuwei Jiang; Yunqi Wang; Ti Zhou
- State Power Investment Corporation (China)
- ?China Investment Corporation (CIC)
- Central University of Finance and Economics
- ?Central University of Finance and Economics (CUFE)
- Southern University of Science and Technology
Strategy in a nutshell
The strategy constructs optimal monthly portfolios for the U.S. and nine other industrialized countries (excluding Japan) using stock market and local risk-free Treasury bill data. The key driver is the U.S. forward variance risk factor (FVF^US), derived from the S&P 500 option-implied volatility (VIX) term structure. Forward variances over multiple horizons (3–6m, 6–9m, 9–12m, 12–18m) are extracted and combined via partial least squares (PLS) regression with one-period-ahead U.S. excess returns as a proxy. The resulting FVF^US factor informs out-of-sample (OOS) forecasts of excess returns for each country. Optimal portfolio weights are calculated using the predicted excess returns, adjusted by forecasted return variance and risk aversion, and portfolios are value-weighted and rebalanced monthly.
Economic rationale
The strategy captures the global influence of U.S. stock market volatility on international equity premiums. The FVF^US factor, reflecting the U.S. forward variance term structure, significantly predicts market returns for the U.S. and nine non-U.S. industrialized countries both in-sample and out-of-sample. Exposure of international equities to U.S. volatility aligns with volatility spillover intensity, demonstrating that U.S. market shocks propagate globally. These findings underscore the importance of incorporating U.S. volatility risk to model conditional international asset pricing and understand cross-country return predictability.