基于改进深度学习的金融时间序列波动率研究  

Research on financial time series volatility based on improved deep learning

作　　者：赵哲玮 ZHAO Zhewei(School of electronic and Information Engineering,Anhui Jianzhu University,Hefei 230601,China)

机构地区：[1]安徽建筑大学电子与信息工程学院,合肥230601

出　　处：《计算机应用文摘》2024年第6期101-104,共4页Chinese Journal of Computer Application

摘　　要：线性模型和传统神经网络模型是常用的传统金融时间序列预测方法,但在非线性、非平稳的金融时间序列预测中存在一定的局限性。对此,文章提出了一种改进的深度学习模型。该模型结合了卷积神经网络和长短时记忆网络,可以有效捕捉金融时间序列中的非线性特征。通过真实数据与预测结果对比发现,文章提出的算法模型有助于捕捉金融时间序列中的非线性特征,可提高波动率预测的准确性。Linear models and traditional neural network models are commonly used traditional financial time series prediction methods,but there are certain limitations in predicting nonlinear and non-stationary financial time series.In this regard,the article proposes an improved deep learning model.This model combines convolutional neural networks and long short-term memory networks,which can effectively capture nonlinear features in financial time series.By comparing real data with predicted results,it was found that the algorithm model proposed in the article helps to capture nonlinear features in financial time series,which can improve the accuracy of volatility prediction.

关 键 词：深度学习 算法优化 金融时间序列 波动率 预测 

分 类 号：TP181[自动化与计算机技术—控制理论与控制工程]