Forecasting volatility with a stacked model based on a hybridized Artificial Neural Network

作者：

Highlights：

• Volatility forecasts are improved by stacking machine learning algorithms.

• Merging Artificial Neural Networks with other models increases its forecasting power.

• Regardless of the volatility model, high volatile regimes lead to higher error rates.

• Risk measurements precision is increased by stacking machine learning models.

摘要

•Volatility forecasts are improved by stacking machine learning algorithms.•Merging Artificial Neural Networks with other models increases its forecasting power.•Regardless of the volatility model, high volatile regimes lead to higher error rates.•Risk measurements precision is increased by stacking machine learning models.

论文关键词：Machine learning,Stacking algorithms,Risk assessment,Volatility forecasting,Hybrid models

论文评审过程：Received 30 October 2018, Revised 26 March 2019, Accepted 27 March 2019, Available online 27 March 2019, Version of Record 2 April 2019.

论文官网地址：https://doi.org/10.1016/j.eswa.2019.03.046