Exploiting intra-day patterns for market shock prediction: A machine learning approach
作者：
Highlights：
• We define market shocks as the innovation of ARMA-GARCH.
• We develop a methodology ARMA-GARCH-NN for market shock prediction.
• A nearest-K cross-validation method is proposed and applied.
• The predictability of market shocks is confirmed by experiments on the S&P 500 data.
• The prediction result is used as a new signal for trading strategy development.
摘要
•We define market shocks as the innovation of ARMA-GARCH.•We develop a methodology ARMA-GARCH-NN for market shock prediction.•A nearest-K cross-validation method is proposed and applied.•The predictability of market shocks is confirmed by experiments on the S&P 500 data.•The prediction result is used as a new signal for trading strategy development.
论文关键词：High-frequency data,Financial forecasting,Neural networks,Time series model
论文评审过程：Received 27 November 2018, Revised 3 March 2019, Accepted 4 March 2019, Available online 6 March 2019, Version of Record 16 March 2019.
论文官网地址：https://doi.org/10.1016/j.eswa.2019.03.006
