Google Trends and Stock Returns

Abstract: The purpose of this thesis is to use social data in the form of Google Trends for companies listed on the S&P 100 to see if they contain information that allows us to predict future returns in the stock market. Using econometric models and trading strategies, the predictive power of Google Trends is investigated both if it can predict future stock returns but also if this knowledge can be used to make arbitrage profits above the market. The results of this study are positive, Google Trends can predict future returns of individual companies and it seems that increases in Google Trends for a company on aggregate predict future negative returns. The study also proves that this knowledge can be used to make arbitrage profits in times of economic stability.