Volatility forecasting with machine learning and intraday commonality

TL;DR

This paper demonstrates that machine learning models, especially neural networks, effectively forecast intraday and next-day realized volatility by leveraging intraday data, market proxies, and commonality across stocks, outperforming traditional methods.

Contribution

It introduces a novel approach combining intraday data pooling and market proxies in machine learning models for volatility forecasting, showing robustness across different stocks and time-of-day effects.

Findings

Neural networks outperform linear and tree-based models in volatility prediction.

Pooling intraday data and market proxies improves forecast accuracy.

Forecasting models generalize well to new stocks not seen during training.

Abstract

We apply machine learning models to forecast intraday realized volatility (RV), by exploiting commonality in intraday volatility via pooling stock data together, and by incorporating a proxy for the market volatility. Neural networks dominate linear regressions and tree-based models in terms of performance, due to their ability to uncover and model complex latent interactions among variables. Our findings remain robust when we apply trained models to new stocks that have not been included in the training set, thus providing new empirical evidence for a universal volatility mechanism among stocks. Finally, we propose a new approach to forecasting one-day-ahead RVs using past intraday RVs as predictors, and highlight interesting time-of-day effects that aid the forecasting mechanism. The results demonstrate that the proposed methodology yields superior out-of-sample forecasts over a strong set of traditional baselines that only rely on past daily RVs.