Title: Expectations, disagreement and news
Authors: Philipp Adaemmer - Helmut Schmidt University (Germany) [presenting]
Joscha Beckmann - University of Duisburg-Essen (Germany)
Rainer Schuessler - Helmut Schmidt University Hamburg (Germany)
Abstract: An increasing amount of research focuses on the effects of news and uncertainty on macroeconomic aggregates. Although it is widely agreed that uncertainty exhibits various transmission channels with regard to the real economy and financial markets, little is known about the effects of economic news on macroeconomic and financial expectations. Recent advances in natural language processing have made it feasible to quantify vast amounts of written texts without relying on pre-determined keywords or manual compilations. We thus combine a correlated topic model and a dictionary based sentiment analysis to extract economic topics from approx. 500,000 U.S. newspaper articles. The results are used to investigate which type of news is correlated with professional economic forecasts and whether this relationship is varying over time. We use a flexible version of dynamic model averaging for the econometric analysis, which allows us to combine a large set of dynamic logistic regression models, differing with respect to the included explanatory variables and the degree of time variation in the parameters. The model's weight within the combination is based on the data support for each individual model, that is, its likelihood. The newspaper articles are obtained from LexisNexis Group and the survey data from Consensus Economics.