Volatility forecasting using deep neural network with time-series feature embedding
Volatility forecasting using deep neural network with time-series feature embedding
复制标题
使用具有时间序列特征嵌入的深度神经网络进行波动率预测
DOI：
10.1080/1331677x.2022.2089192
复制
发表时间：
2022-06
期刊：
影响因子：
--
通讯作者：
Yuan-Hai Shao
中科院分区：
文献类型：
--
作者：
Wei-Jie Chen;Jing-Jing Yao;Yuan-Hai Shao
摘要波动性通常是市场变化或趋势的代理指标，包含了投资者和政策制定者的基本信息。提出了一种用于波动率预测的时间嵌入混合深度神经网络模型(HDNN)。我们的HDNN的主要思想是将一维时间序列数据编码为二维GAF图像，这使得后续的卷积神经网络(CNN)能够自动学习与波动相关的特征映射。具体地说，HDNN采用了一种优雅的端到端学习范式来进行波动率预测，该范式由特征嵌入和回归组件组成。特征嵌入组件通过在底层时间嵌入空间中精心设计的CNN来探索来自GAF图像的与波动性相关的时间信息。然后，回归组件将这些嵌入向量作为波动率预测任务的输入。最后，我们在四个合成的GBM数据集和五个真实的股票指数数据集上，从五个回归度量的角度检验了HDNN的可行性。结果表明，HDNN在大多数情况下比GARCH、EGACH、SVR和NN等基线预测模型具有更好的性能。证实了HDNN提取的波动率相关时间特征确实提高了预测能力。此外，Friedman检验验证了HDNN在统计上优于比较的预测模型。
Abstract Volatility is usually a proxy indicator for market variation or tendency, containing essential information for investors and policy-makers. This paper proposes a novel hybrid deep neural network model (HDNN) with temporal embedding for volatility forecasting. The main idea of our HDNN is that it encodes one-dimensional time-series data as two-dimensional GAF images, which enables the follow-up convolution neural network (CNN) to learn volatility-related feature mappings automatically. Specifically, HDNN adopts an elegant end-to-end learning paradigm for volatility forecasting, which consists of feature embedding and regression components. The feature embedding component explores the volatility-related temporal information from GAF images via the elaborate CNN in an underlying temporal embedding space. Then, the regression component takes these embedding vectors as input for volatility forecasting tasks. Finally, we examine the feasibility of HDNN on four synthetic GBM datasets and five real-world Stock Index datasets in terms of five regression metrics. The results demonstrate that HDNN has better performance in most cases than the baseline forecasting models of GARCH, EGACH, SVR, and NN. It confirms that the volatility-related temporal features extracted by HDNN indeed improve the forecasting ability. Furthermore, the Friedman test verifies that HDNN is statistically superior to the compared forecasting models.
登录
查看更多内容
影响因子：
--
作者：
Dvorsky, Jan;Belas, Jaroslav;Brabenec, Tomas
通讯作者：
Brabenec, Tomas
影响因子：
5.8
作者：
Taoying Li;Yuqi Zhang;Tingniao Wang
通讯作者：
Taoying Li;Yuqi Zhang;Tingniao Wang
DOI：
10.1016/j.ejor.2021.02.033
发表时间：
2021-02
期刊：
Eur. J. Oper. Res.
影响因子：
--
作者：
Lorenzo Lampariello;C. Neumann;J. M. Ricci;Simone Sagratella;O. Stein
通讯作者：
Lorenzo Lampariello;C. Neumann;J. M. Ricci;Simone Sagratella;O. Stein
DOI：
10.1016/j.najef.2019.101027
发表时间：
2019-07
期刊：
The North American Journal of Economics and Finance
影响因子：
--
作者：
Sasmita Mishra;S. Padhy
通讯作者：
Sasmita Mishra;S. Padhy
DOI：
10.1016/j.eswa.2019.03.046
发表时间：
2019-09
期刊：
Expert Syst. Appl.
影响因子：
--
作者：
Eduardo Ramos-Pérez;P. Alonso-González;J. J. Núñez-Velázquez-J.
通讯作者：
Eduardo Ramos-Pérez;P. Alonso-González;J. J. Núñez-Velázquez-J.