Enhancing Stock Market Prediction with Extended Coupled Hidden Markov Model over Multi-Sourced Data

Noname manuscript No. (will be inserted by the editor) Abstract Traditional stock market prediction methods commonly only utilize the historical trading data, ignoring the fact that stock market fluctuations can be impacted by various other information sources such as stock related events. Although some recent works propose event-driven prediction approaches by considering the event data, how to leverage the joint impacts of multiple data sources still remains an open research problem. In this work, we study how to explore multiple data sources to improve the performance of the stock prediction. We introduce an Extended Coupled Hidden Markov Model incorporating the news events with the historical trading data. To address the data sparsity issue of news events for each single stock, we further study the fluctuation correlations between the stocks and incorporate the correlations into the model to facilitate the prediction task. Keywords Stock prediction · Event extraction · Information fusion · Hidden Markov Model 1 Introduction The capability of predicting the stock price movement directions can offer enormous arbitrage profit opportunities and thus attract much attention from both academia and industry. Conventional quantitative trading prediction methods are mostly based on the historical trading data such as prices and volumes. According to the Efficient Market Hypothesis (EMH) [16], stock prices are the reflection of all known information. Key Laboratory of Trustworthy Distributed Computing and Service (Beijing University of Posts and Telecommunications), Ministry of Education, Beijing, China. As more and more investors obtain information from social media [49, 57], the indicators obtained from Web news articles and social networks can also have significant impacts on the stock prices, and thus such factors that can derive the stock price fluctuations must be considered. As such, there are growing research interests in exploring financial text documents such as news articles, financial standings to facilitate the stock prediction task.