Predicting financial volatility: High‐frequency time‐series forecasts vis‐à‐vis implied volatility
利用高频数据和长记忆模型改进金融波动的测量与预测，发现基于历史日内收益的波动率预测可与隐含波动率竞争甚至超越，研究覆盖股票、外汇和商品三类资产。
Abstract Recent evidence suggests option implied volatilities provide better forecasts of financial volatility than time‐series models based on historical daily returns. In this study both the measurement and the forecasting of financial volatility is improved using high‐frequency data and long memory modeling, the latest proposed method to model volatility. This is the first study to extract results for three separate asset classes, equity, foreign exchange, and commodities. The results for the S&P 500, YEN/USD, and Light, Sweet Crude Oil provide a robust indication that volatility forecasts based on historical intraday returns do provide good volatility forecasts that can compete with and even outperform implied volatility. © 2004 Wiley Periodicals, Inc. Jrl Fut Mark 24:1005–1028, 2004