Publication

A Rational Inattention Perspective on Equilibrium Asset Pricing under Heterogeneous Information with Structural Breaks and Market Efficiency

Jan 1, 2016 · 2 authors · 7 topics

Authors

Steve HeinkeWarmuth Niels

Topics

Complex Systems and Time Series AnalysisFinancial Markets and Investment StrategiesEconomic theories and models2016By Steve Heinke and Niels J. Warmuth∗In this paper we present a new model of how information travels within financial markets and present empirical evidence that the concept of attention driven information efficiency is more conju gate with market data as compared to the prevailing concept of efficient markets. Augmenting our model by a shift component made it possible to explain shifts in asset prices by a lack of atten tion on small permanent changes in the fundamentals. This can also be seen as a micro-level explanation of the momentum effect. By a further augmentation of the model through the introduction of heterogeneous information processing capacities we are able to give a fundamental interpretation of the financial services indus try as providers of information processing capacity. Moreover, the burst of the housing bubble in the US and the successful bet of John Paulson against it are shown to be prime empirical examples of our framework. JEL: C60, D80, D83, D84, D92, G12, G14 Keywords: Limited attention, asset pricing, rational inattention, momentum trading∗ Heinke, Steve: University of Zurich, Wintherturerstrasse 30 - 8057 Zurich, steve.heinke@econ.uzh.ch. Warmuth, Niels Jakob: Munich University of Technology, Arcisstrasse 21 - 80333 Munich, niels.warmuth@wi.tum.de. Acknowledgements: We are grateful to Michael Allman, Marios Angele tos, Andreas Hefti, Josef Falkinger, Igor Letina, Sabrina Studer, Thomas Schmid, Christoph J¨ackel, Christoph Kaserer, and to seminar participants at the Munich University of Technology, University of Zurich, and the Zurich Workshop on Economics 2013 in Schaffhausen (Switzerland). 1

About

PublishedJan 1, 2016
TypeReport
Citations0

Powered by the Exa API