摘要
本文引入互联网财经新闻信息来对日汇率波动趋势进行更准确预测。使用深度学习文本分析模型提取新闻中与汇率波动趋势有关的信息,计算日新闻影响数值特征,并融合汇率历史数据建立时间卷积网络模型对汇率波动趋势进行预测,最后根据预测结果研究交易策略并进行模拟投资。研究新闻对汇率预测的影响作用可以为投资决策提供实证依据与理论支撑,为新闻与汇率波动建立了联系。实证表明,该方法能提高汇率波动趋势预测的准确率并能获得较高的投资收益,体现了新闻信息对汇率波动预测的重要影响作用。
To study the accurate forecast of daily exchange rate fluctuations, this paper introduces information from Internet financial news.The deep learning text analysis model is used to extract information related to the trend of exchange rate fluctuations in news, calculate the numerical characteristics of the influence of daily news, in addition,build a Temporal Convolutional Net work model based on historical exchange rate data to predict the trend of exchange rate fluctuations.Finally, trading strategics are studied and simulated investments arc made based on the predicted results.The research on the influence of news on exchange rate prediction can provide empirical basis and theoretical support for investment decisions and establish a link between news and exchange rate fluctuations.The empirical results show that this method can improve the accuracy of the prediction of the trend of exchange rate fluctuations and obtain higher investment returns,which reflects the important influence of news information on the prediction of exchange rate fluctuations.
作者
张杰
张永卿
翟东升
ZHANG Jie;ZHANG Yong-qing;ZHAI Dong-sheng(School of Economics and Management,Beijing University of Technology,Beijing 100124,China)
出处
《系统工程》
北大核心
2021年第3期121-131,共11页
Systems Engineering
关键词
文本挖掘
汇率波动预测
深度学习
新闻信息
Text Mining
Forecast of Exchange Rate Fluctuations
Deep Learning
News Information
