Volatility forecasting for low-volatility investing
研究了用多种波动率模型预测美国500只大股票的波动率，并据此构建低波动率投资组合，发现基于面板异质自回归模型和预测组合的策略在扣除交易成本后表现最佳，且易于实时实施。
Low-volatility investing often involves sorting and selecting stocks based on retrospective risk measures, for example, the historical standard deviation of returns. In contrast, we employ volatility forecasts from various volatility models to sort, select, and estimate portfolio weights on the 500 largest US stocks. We find that exploiting a large set of time-series models delivers large, significant economic gains compared to traditional benchmarks. After accounting for transaction costs, a low-volatility portfolio based on volatility forecasts from a panel heterogeneous autoregression model and a portfolio based on forecast combinations perform best and can be easily implemented in real time.