 Improving stock price prediction using the long short-term memory model combined with online social networksKeyan Liu, Jianan Zhou and Dayong Dong Journal of Behavioral and Experimental Finance, 2021, vol. 30, issue C Abstract: To predict stock prices with effective information has always been a problem of great significance in the fields of behavioral finance. In this paper, we predict the stock prices with novel online data sources. For some emerging countries (such as China), individual investors often obtain trading information from online social media platforms. Therefore, stock features extracted from social media platforms are likely to include valuable information. We obtained the data of users and stocks they followed from EastyMoney, China’s largest social media platform, and generate daily social networks. Then we calculated the network variable of each stock as a supplement to traditional variables, and predicted the close prices of the SSE 50 constituent stocks using the LSTM model. The empirical results show that the social network variable can effectively improve the prediction accuracy. Our results can help investors improve forecasting accuracy. Our findings can help investors enhancing the understanding of the link between social networks and stock prices. Keywords: Online social networks; LSTM; Stock prices (search for similar items in EconPapers) Downloads: (external link) Related works: Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text Persistent link: https://EconPapers.repec.org/RePEc:eee:beexfi:v:30:y:2021:i:c:s2214635021000514 DOI: 10.1016/j.jbef.2021.100507 Access Statistics for this article Journal of Behavioral and Experimental Finance is currently edited by Michael Dowling and Jürgen Huber More articles in Journal of Behavioral and Experimental Finance from Elsevier |
Is your work missing from RePEc? Here is how to contribute.
Questions or problems? Check the EconPapers FAQ or send mail to .
EconPapers is hosted by the
School of Business at Örebro University.