The role of attribute selection in Deep ANNs learning framework for high-frequency financial trading

Q1 Economics, Econometrics and Finance

Intelligent Systems in Accounting, Finance and Management Pub Date : 2020-03-12 DOI:10.1002/isaf.1466

Monira Essa Aloud

{"title":"The role of attribute selection in Deep ANNs learning framework for high-frequency financial trading","authors":"Monira Essa Aloud","doi":"10.1002/isaf.1466","DOIUrl":null,"url":null,"abstract":"<div>\n \n <p>In financial trading, technical and quantitative analysis tools are used for the development of decision support systems. Although these traditional tools are useful, new techniques in the field of machine learning have been developed for time-series forecasting. This paper analyses the role of attribute selection on the development of a simple deep-learning ANN (D-ANN) multi-agent framework to accomplish a profitable trading strategy in the course of a series of trading simulations in the foreign exchange market. The paper evaluates the performance of the D-ANN multi-agent framework over different time spans of high-frequency (HF) intraday asset time-series data and determines how a set of the framework attributes produces effective forecasting for profitable trading. The paper shows the existence of predictable short-term price trends in the market time series, and an understanding of the probability of price movements may be useful to HF traders. The results of this paper can be used to further develop financial decision-support systems and autonomous trading strategies for the financial market.</p>\n </div>","PeriodicalId":53473,"journal":{"name":"Intelligent Systems in Accounting, Finance and Management","volume":"27 2","pages":"43-54"},"PeriodicalIF":0.0000,"publicationDate":"2020-03-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://sci-hub-pdf.com/10.1002/isaf.1466","citationCount":"3","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Intelligent Systems in Accounting, Finance and Management","FirstCategoryId":"1085","ListUrlMain":"https://onlinelibrary.wiley.com/doi/10.1002/isaf.1466","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"Economics, Econometrics and Finance","Score":null,"Total":0}

引用次数: 3

Abstract

In financial trading, technical and quantitative analysis tools are used for the development of decision support systems. Although these traditional tools are useful, new techniques in the field of machine learning have been developed for time-series forecasting. This paper analyses the role of attribute selection on the development of a simple deep-learning ANN (D-ANN) multi-agent framework to accomplish a profitable trading strategy in the course of a series of trading simulations in the foreign exchange market. The paper evaluates the performance of the D-ANN multi-agent framework over different time spans of high-frequency (HF) intraday asset time-series data and determines how a set of the framework attributes produces effective forecasting for profitable trading. The paper shows the existence of predictable short-term price trends in the market time series, and an understanding of the probability of price movements may be useful to HF traders. The results of this paper can be used to further develop financial decision-support systems and autonomous trading strategies for the financial market.

查看原文本刊更多论文

属性选择在深度人工神经网络高频金融交易学习框架中的作用

在金融交易中，技术和定量分析工具用于决策支持系统的开发。虽然这些传统的工具是有用的，但机器学习领域的新技术已经被开发出来用于时间序列预测。在外汇市场的一系列交易模拟过程中，分析了属性选择在开发简单深度学习人工神经网络(D-ANN)多智能体框架以实现盈利交易策略中的作用。本文评估了D-ANN多智能体框架在高频(HF)日内资产时间序列数据的不同时间跨度上的性能，并确定了一组框架属性如何为有利可图的交易产生有效的预测。本文表明，在市场时间序列中存在可预测的短期价格趋势，对价格变动概率的理解可能对高频交易者有用。本文的研究结果可用于进一步开发金融市场的金融决策支持系统和自主交易策略。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

Intelligent Systems in Accounting, Finance and Management Economics, Econometrics and Finance-Finance

CiteScore

6.00

自引率

0.00%

发文量

期刊介绍： Intelligent Systems in Accounting, Finance and Management is a quarterly international journal which publishes original, high quality material dealing with all aspects of intelligent systems as they relate to the fields of accounting, economics, finance, marketing and management. In addition, the journal also is concerned with related emerging technologies, including big data, business intelligence, social media and other technologies. It encourages the development of novel technologies, and the embedding of new and existing technologies into applications of real, practical value. Therefore, implementation issues are of as much concern as development issues. The journal is designed to appeal to academics in the intelligent systems, emerging technologies and business fields, as well as to advanced practitioners who wish to improve the effectiveness, efficiency, or economy of their working practices. A special feature of the journal is the use of two groups of reviewers, those who specialize in intelligent systems work, and also those who specialize in applications areas. Reviewers are asked to address issues of originality and actual or potential impact on research, teaching, or practice in the accounting, finance, or management fields. Authors working on conceptual developments or on laboratory-based explorations of data sets therefore need to address the issue of potential impact at some level in submissions to the journal.