A multi-view time series model for share turnover prediction

Share turnover is a key indicator for investing in the stock market, which represents how easy or difficult it is to trade a stock. Several techniques have been proposed to predict share turnover values. However, they are often inaccurate because they utilize single-view models that have an incomple...

Full description

Saved in:
Bibliographic Details
Published in:Applied intelligence (Dordrecht, Netherlands) Netherlands), 2022-10, Vol.52 (13), p.14595-14606
Main Authors: Wang, Zhijin, Su, Qiankun, Chao, Guoqing, Cai, Bing, Huang, Yaohui, Fu, Yonggang
Format: Article
Language:English
Subjects:
Algorithms
Artificial Intelligence
Computer Science
Machines
Manufacturing
Mechanical Engineering
Processes
Special Issue on Multi-view Learning
Time series
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Share turnover is a key indicator for investing in the stock market, which represents how easy or difficult it is to trade a stock. Several techniques have been proposed to predict share turnover values. However, they are often inaccurate because they utilize single-view models that have an incomplete picture of the temporal dynamics. To address this issue, a multi-view time series model (MvT) is proposed to capture temporal dynamics using three views on two data groups. The temporal dynamics of target turnover data and exogenous turnover data are captured by a view generation component. The component generates three views in three different aspects. The predictions are then made by a view combination component and a full connected layer. Extensive experiments on two real stock datasets show the effectiveness and efficiency of the proposed MvT model, when compared with ten algorithms on four groups of stock data in terms of three metrics.
ISSN:0924-669X
1573-7497
DOI:10.1007/s10489-021-02979-y