Coupling News Sentiment with Web Browsing Data Improves Prediction of Intra-Day Price Dynamics

The new digital revolution of big data is deeply changing our capability= of understanding society and forecasting the outcome of many social and ec= onomic systems. Unfortunately, information can be very heterogeneous in the= importance, relevance, and surprise it conveys, affecting severely the pre= dictive power of semantic and statistical methods. Here we show that the ag= gregation of web users' behavior can be elicited to overcome this problem i= n a hard to predict complex system, namely the financial market. Specifical= ly, our in-sample analysis shows that the combined use of sentiment analysi= s of news and browsing activity of users of Yahoo! Finance greatly helps fo= recasting intra-day and daily price changes of a set of 100 highly capitali= zed US stocks traded in the period 2012-2013. Sentiment analysis or browsin= g activity when taken alone have very small or no predictive power. Convers= ely, when considering a news signal where in a given time interval we compu= te the average sentiment of the clicked news, weighted by the number of cli= cks, we show that for nearly 50% of the companies such signal Granger-cause= s hourly price returns. Our result indicates a "wisdom- of-the-crowd" effec= t that allows to exploit users' activity to identify and weigh properly the= relevant and surprising news, enhancing considerably the forecasting power= of the news sentiment.