Quantifying correlation between Financial News and stocks

2016 IEEE Symposium Series on Computational Intelligence (SSCI) Pub Date : 2016-12-01 DOI:10.1109/SSCI.2016.7850021

Haizhou Qu, D. Kazakov

{"title":"Quantifying correlation between Financial News and stocks","authors":"Haizhou Qu, D. Kazakov","doi":"10.1109/SSCI.2016.7850021","DOIUrl":null,"url":null,"abstract":"Financial news and stocks appear linked to the point where the use of online news to forecast the markets has become a major selling point for some traders. The correlation between news content and stock returns is clearly of interest, but has been mostly centred on news meta-data, such as volume and popularity. We address this question here by measuring the correlation between the returns of 27 publicly traded companies and news about them as collected from Yahoo Financial News for the period 1 Oct 2014 to 30 Apr 2015. In all reported experiments, two metrics are defined, one to measure the distance between two time series, the other to quantify the difference between two collections of news items. Two 27 × 27 distance matrices are thus produced, and their correlation measured with the Mantel test. This allows us to estimate the correlation of stock market data (returns, change, volume and close price) with the content of published news in a given period of time. A number of representations for the news are tested, as well as different distance metrics between time series. Clear, statistically significant, moderate level correlations are detected in most cases. Lastly, the impact of the length of the period studied on the observed correlation is also investigated.","PeriodicalId":120288,"journal":{"name":"2016 IEEE Symposium Series on Computational Intelligence (SSCI)","volume":null,"pages":null},"PeriodicalIF":0.0000,"publicationDate":"2016-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"1","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2016 IEEE Symposium Series on Computational Intelligence (SSCI)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/SSCI.2016.7850021","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}

引用次数: 1

Abstract

Financial news and stocks appear linked to the point where the use of online news to forecast the markets has become a major selling point for some traders. The correlation between news content and stock returns is clearly of interest, but has been mostly centred on news meta-data, such as volume and popularity. We address this question here by measuring the correlation between the returns of 27 publicly traded companies and news about them as collected from Yahoo Financial News for the period 1 Oct 2014 to 30 Apr 2015. In all reported experiments, two metrics are defined, one to measure the distance between two time series, the other to quantify the difference between two collections of news items. Two 27 × 27 distance matrices are thus produced, and their correlation measured with the Mantel test. This allows us to estimate the correlation of stock market data (returns, change, volume and close price) with the content of published news in a given period of time. A number of representations for the news are tested, as well as different distance metrics between time series. Clear, statistically significant, moderate level correlations are detected in most cases. Lastly, the impact of the length of the period studied on the observed correlation is also investigated.

查看原文本刊更多论文

量化财经新闻与股票的相关性

金融新闻和股票似乎联系在一起，利用在线新闻预测市场已成为一些交易员的主要卖点。新闻内容与股票回报之间的相关性显然令人感兴趣，但主要集中在新闻元数据上，如数量和受欢迎程度。我们在这里通过测量27家上市公司的收益与从雅虎财经新闻收集的2014年10月1日至2015年4月30日期间有关它们的新闻之间的相关性来解决这个问题。在所有报告的实验中，定义了两个指标，一个用于测量两个时间序列之间的距离，另一个用于量化两个新闻项目集合之间的差异。由此产生了两个27 × 27的距离矩阵，并用Mantel测试测量了它们的相关性。这使我们能够估计给定时间段内股票市场数据(收益、变动、成交量和收盘价)与发布的新闻内容之间的相关性。测试了新闻的许多表示形式，以及时间序列之间的不同距离度量。在大多数情况下，检测到明确的、统计上显著的、中等水平的相关性。最后，还研究了研究周期长度对观测到的相关性的影响。

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

2016 IEEE Symposium Series on Computational Intelligence (SSCI)

自引率

0.00%

发文量