Exploring Relationship between Headline News Sentiment and Stock Return

Information from news is an integral part of investment activity. Such information is widely available from online media platforms, in which we can obtain relatively easy thanks to the development of technology. Despite predicting future stock prices being a hard challenge, the availability of big data is able to support investment decisions. Any relevant news is deemed as a determinant factor for influencing the public decision making. The Efficient Market Hypothesis states that financial markets depend on the availability of information. Sentiment analysis act as assisting tool to support the sentiment classification process, which falls into positive or negative categories. The process uses Naive Bayes and Support Vector Machine method. We conduct the research to see the relationship of headlines news from 20 companies listed in the Indonesia LQ45 index (February 2013 - August 2018) and stock returns. This study is carried out by Spearman rank correlation in looking at the relationship between stock returns and headline news. The findings of the study indicates that there is a correlation between headline news and stock return.