Web Sentiment Analysis for Revealing Public Opinions, Trends and Making Good Financial Decisions

This study attempts to discover and analyze the predictive power of stock messages, posting on financial message boards, on future stock price directional movements. We construct a set of robust models based on sentiment analysis and data mining algorithms. Our dataset consist of 447,393 messages, on the 30 Dow Jones Index (DJIA) stocks, posted on the Yahoo! Finance message board in the period August 2012 to May 2013, of which 55,217 with sentiment tag and 5,967 distinct authors. We propose a novel way to generate sentiment based on author’s credibility, calculated on accuracy of his past messages. Our results provide empirical evidence that, using our method (3 and 5 scale index models), there is strong and useful information on financial message boards pertinent to stock market movements. In addition, we demonstrate that we can use this information in order to make accurate predictions about the return on investment and to implement good trading strategies based on sentiment analysis, doing, on average, much better than traditional investment strategies like Buy and Hold or Moving Averages (5-20 periods). Our results appear to be statistically and economically significant. Theory that suggests a link between small investor behavior and stock market performance is now supported by our work.