Big data analytics for financial Market volatility forecast based on support vector machine

Highlights：

• The realized volatility and the realized bi-power variation have the obvious phenomenon of fluctuation aggregation, and have the auto-correlation; different from the auto-correlation of logarithmic yield, the auto-correlation of the realized volatility and the realized bi-power variation is relatively strong, and the correlation is positive.

• The verification results of verification data obtained by fitting HAR-RV model, HAR-lnRV model and HAR-JV-CV model show that: the HAR-lnRV model has the best prediction effect, followed by the HAR-JV-CV model, and the worst is the HAR-RV model.

摘要

•Volatility is an important measurement index of market risk, and the research and forecasting on the volatility of high-frequency data is of great significance to investors, government regulators and capital markets.•The realized volatility and the realized bi-power variation have the obvious phenomenon of fluctuation aggregation, and have the auto-correlation; different from the auto-correlation of logarithmic yield, the auto-correlation of the realized volatility and the realized bi-power variation is relatively strong, and the correlation is positive.•The verification results of verification data obtained by fitting HAR-RV model, HAR-lnRV model and HAR-JV-CV model show that: the HAR-lnRV model has the best prediction effect, followed by the HAR-JV-CV model, and the worst is the HAR-RV model.

论文评审过程：Received 26 December 2018, Revised 13 May 2019, Accepted 30 May 2019, Available online 13 June 2019, Version of Record 21 November 2019.