Recovering Heterogeneous Beliefs and Preferences from Asset Prices

48 Pages Posted: 20 Jul 2020

See all articles by Anisha Ghosh

Anisha Ghosh

McGill University

Arthur G. Korteweg

University of Southern California - Marshall School of Business

Qing Xu

McGill University, Desautels Faculty of Management, Students

Date Written: July 12, 2020

Abstract

We propose a novel information-theoretic approach to separately identify the risk preferences and beliefs of different types of financial market investors. Investors who allocate most of their wealth in large market capitalization stocks are risk averse and believe that the aggregate stock market return is strongly countercyclical. In contrast, investors in small-growth stocks are substantially less risk averse and believe in procyclical expected stock market returns. Our findings can reconcile the procyclical expected market returns found in investor survey data with the countercyclical expected returns implied by rational expectations models.

Keywords: asset pricing, beliefs, preferences, heterogeneity, business cycle, rational expectations, behavioral finance, smoothed empirical likelihood

JEL Classification: C51, E3, E70, G12, G14, G40

Suggested Citation

Ghosh, Anisha and Korteweg, Arthur G. and Xu, Qing, Recovering Heterogeneous Beliefs and Preferences from Asset Prices (July 12, 2020). Available at SSRN: https://ssrn.com/abstract=3650199 or http://dx.doi.org/10.2139/ssrn.3650199

Anisha Ghosh

McGill University ( email )

1001 Sherbrooke St. W
Montreal, Quebec H3A 1G5
Canada

Arthur G. Korteweg (Contact Author)

University of Southern California - Marshall School of Business ( email )

3670 Trousdale Parkway
Los Angeles, CA 90089
United States

HOME PAGE: http://www.marshall.usc.edu/personnel/arthur-korteweg

Qing Xu

McGill University, Desautels Faculty of Management, Students ( email )

1001 Sherbrooke St. West
Montreal
Canada

Do you have a job opening that you would like to promote on SSRN?

Paper statistics

Downloads
105
Abstract Views
399
rank
308,728
PlumX Metrics