The distinct impact of attention and news sentiment on stock prices

Abstract

In recent years, there has been an increasing interest in analyzing the impact of information contained in news articles on financial markets. The reaction effect of stock markets to the news’ underlying sentiment has been demonstrated in the literature. This thesis aims at measuring with statistical means if attention is a driving element for sentiment to impact the stock market more consistently. A Panel Vector Autoregression analysis over the year 2016 on the constituents of the S&P 100 is conducted to assess this impact. I use indirect and direct proxies of attention combined with sentiment found in news, to create an interaction measure. The results show that a one-week lagged impact of sentiment on abnormal return is greater whenever attention is high. These are consistent with the conceptual assumption that attention is needed for any type of information to influence stock markets and has quantitatively been proven in this work.