Thursday, August 1, 2013

Anomaly: Newspaper Content Can Predict Stock Market Returns

Anybody wanna buy some low latency machine readable feeds? It appears good 'ol word counts work just as well.
From the University of St. Gallen - Swiss Institute of Banking and Finance:

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. Word-count 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.

Number of Pages in PDF File: 40
Keywords: Wordcount, Text Mining, Market Efficiency, Tactical Asset Allocation
JEL Classification: G10
working papers series

Free Download

This is one of the older papers the University of St. Gallen is contributing to the SSRN, here are a couple of others:
Has Hedge Fund Alpha Disappeared?
Price Discovery and Information Transmission Among Asset Markets: A High-Frequency Perspective