Forecasting the Equity Risk Premium: The Importance of Regime-Dependent Evaluation

49 Pages Posted: 19 Jan 2016 Last revised: 22 Sep 2020

See all articles by Nick Baltas

Nick Baltas

Imperial College Business School; Goldman Sachs International

Dimitris Karyampas

Bocconi University

Date Written: September 25, 2017

Abstract

Asset allocation is critically dependent on the ability to forecast the equity risk premium (ERP) out-of-sample. But, is superior econometric predictability across the business cycle synonymous to predictability at all times? We evaluate recently introduced ERP forecasting models, which have been shown to generate econometrically superior ERP forecasts, and find that their forecasting ability is regime-dependent. They give rise to significant relative losses during market downturns, when it matters the most for asset allocators to retain assets and their client base intact. Conversely, any economic benefit occurring during market upswings is diminished for high risk averse and leverage constrained investors.

Keywords: quity risk premium, predictability, out-of-sample forecasting, economic constraints, predictive regression, asset allocation, business cycles, profitability, trading strategies

JEL Classification: C53, C58, G11, G14, G17

Suggested Citation

Baltas, Nick and Baltas, Nick and Karyampas, Dimitris, Forecasting the Equity Risk Premium: The Importance of Regime-Dependent Evaluation (September 25, 2017). Journal of Financial Markets, Volume 38, March 2018, Pages 83-102, Available at SSRN: https://ssrn.com/abstract=2717706 or http://dx.doi.org/10.2139/ssrn.2717706

Nick Baltas (Contact Author)

Imperial College Business School ( email )

South Kensington Campus
Exhibition Road
London, SW7 2AZ
United Kingdom

Goldman Sachs International

Peterborough Court
133 Fleet Street
London, EC4A 2BB
United Kingdom

Dimitris Karyampas

Bocconi University ( email )

Via Sarfatti, 25
Milan, MI 20136
Italy

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

Paper statistics

Downloads
801
Abstract Views
3,145
rank
43,812
PlumX Metrics