Do Newspaper Articles Predict Aggregate Stock Returns?

Abstract

We analyze whether newspaper content can predict aggregate future stock returns. Our study is based on articles published in the Handelsblatt, a leading German financial newspaper, from July 1989 to March 2011. We summarize newspaper content in a systematic way by constructing word-count indices for a large number of words. Wordcount indices are instantly available and therefore potentially valuable financial indicators. Our main finding is that the predictive power of newspaper content has increased over time, particularly since 2000. We find that a cluster analysis approach increases the predictive power of newspaper articles substantially. To obtain optimal predictive power, we need at least seven clusters. Our analysis shows that newspaper content is a valuable predictor of future DAX returns in and out of sample.

Story beneath story: Do magazine articles reveal forthcoming returns on stock market?

“…Nonetheless, it proved useful in predicting forthcoming stock market returns in some previous studies (Ammann et al, 2014;Barber and Loeffler, 1993;Fang et al, 2014;Tetlock et al, 2008).…”

“…In this regard, main reference is Ammann et al (2014) who summarized articles published in Handelsblatt, a leading financial newspaper, by constructing word count indices and found them valuable predictors of future DAX (German Stock Exchange) returns, both in and out of sample. They extended the study with cluster analysis and showed that optimal level of fragmentation of news content, for best predictive values, was seven clusters.…”

“…Our model has inherited most of its features from the study of Ammann et al (2014). Because of the highly multivariate nature of the data, we employ stepwise regression which is an automated tool, having the ease of acting as a single model.…”

“…We mostly rely on control variables evidenced to have forecasting power in previous work (Schrimpf et al, 2007;Walkshaeusel and Lobe, 2011;Ammann et al 2014). As one would expect, these control variables have been ascertained to exhibit a forecasting effect.…”

Using Newspapers for Tracking the Business Cycle: A Comparative Study for Germany and Switzerland

“…Ammann et al (2013) summarise the content of newspaper articles by means of word-count indices and use them to predict DAX 1 stock returns in and out of sample. Engelberg and Parsons (2011) show that press coverage is significantly correlated with the daily trading volume of retail investors.…”