Neural network predictions of the high-frequency CSI300 first distant futures trading volume

Xiaojie Xu () and Yun Zhang ()
Additional contact information
Xiaojie Xu: North Carolina State University
Yun Zhang: North Carolina State University

Financial Markets and Portfolio Management, 2023, vol. 37, issue 2, No 3, 207 pages

Abstract: Abstract Predictions of financial index trading volumes represent an essential issue to market participants and policy makers. We investigate this problem for the high-frequency one-minute CSI300 first distant futures trading volume from the launch of the futures index to two years after all its constituent stocks becoming shortable, a period with generally continuously expanding trading magnitude. We utilize the neural network to model this complex and irregular trading volume series and attempt to answer the following questions: can the trading volume be predicted by its own lags, and if so, how far to predict and how well; can the nearby futures or spot trading volume series help predictions, and if so, by how much; how complex does the model need to be and how robust can it be? We find that the trading volume could be predicted using one to thirty minutes ahead data through a relatively low complex model with five hidden neurons based on its own lags, resulting in generally accurate and stable performance based on the relative root mean square error. Incorporating the nearby futures or spot trading volume series generally does not help improve predictions. Results here should be of interest and use to design trading platforms, monitor system risk, and form index price predictions.

Keywords: CSI300 futures; CSI300 spot; Trading volume; Forecasting; Neural network; C22; C45; G12; G15 (search for similar items in EconPapers)
Date: 2023
References: Add references at CitEc
Citations:

Downloads: (external link)
http://link.springer.com/10.1007/s11408-022-00421-y Abstract (text/html)
Access to full text is restricted to subscribers.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:kap:fmktpm:v:37:y:2023:i:2:d:10.1007_s11408-022-00421-y

Ordering information: This journal article can be ordered from
http://www.springer. ... nt/journal/11408/PS2

DOI: 10.1007/s11408-022-00421-y

Access Statistics for this article

Financial Markets and Portfolio Management is currently edited by Manuel Ammann

More articles in Financial Markets and Portfolio Management from Springer, Swiss Society for Financial Market Research Contact information at EDIRC.
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().