Logistics forum based prediction on stock index using intelligent data analysis and processing of online web posts
- PDF / 1,072,865 Bytes
- 10 Pages / 595.276 x 790.866 pts Page_size
- 116 Downloads / 156 Views
ORIGINAL RESEARCH
Logistics forum based prediction on stock index using intelligent data analysis and processing of online web posts Jinghua Zhao1 · Na Sun1 · Wanyun Cheng2 Received: 26 June 2019 / Accepted: 21 September 2019 © Springer-Verlag GmbH Germany, part of Springer Nature 2019
Abstract This paper focuses on the e-commerce contents of logistics forum. We construct four logistics forums metrics on the degree of busyness and emotional states of the logistics staffs based through hypertext analysis techniques. By choosing a set of individual stocks from e-commerce sector and establishing e-commerce stock index, we examine the empirical evidence of the predictability and degree of influence of e-commerce logistics forum on e-commerce stock index. We find that emotion index and size of web posts may help predict ratio of earnings whereas emotion index, web post index, and disagreement index may help predict exchange volumes. Meanwhile, logistics forum metrics have short term or mid-term influence on e-commerce stock index, ratio of earnings, volatility, and trading volume whereas the influence of emotion index could last more than 40 weeks. Our results help people to understand the interrelationship between the forum contents of logistics social network and stock market activities and provide important decision advice for investors. Keywords Logistics Forum · Online posts · e-Commerce · Stock Market · Predictability · Hypertext analysis
1 Introduction In the past decade, text analysis had been widely used in financial market. By applying text analysis on the contents of social networking media, analysts can get rich verbatim information. Such verbatim information provides an effective tool for understanding basic information, evaluating future profitability, and predicting performance on a company. There are three main research areas on the relationship between social networking media and stock market performance based on the data sources: the public releases of the companies and reports of third party professional analysts, information exchange and discussion among individual
* Na Sun [email protected] Jinghua Zhao [email protected] Wanyun Cheng [email protected] 1
School of Business, University of Shanghai for Science and Technology, Shanghai, China
Shanghai Stock Exchange, Shanghai, China
2
investors, and information exchange and discussion among participants of supply chains. The first data sources include periodic and non-periodic reports of the stock market companies, analyst’s reports and news coverage, etc. researchers found the verbatim extraction (e.g. positive and negative words, phase and attitudes, etc.) can provide effective investment information for investors. However, reports from stock companies are usually not contemporary, analyst reports may have bias due to analyst’s own personal favor, news correspondences cannot be long term stable source of information due to the diversity of news coverage though it usually provides relatively high dependable, real-time and ob
Data Loading...
