Score-Driven Asset Pricing: Predicting Time-Varying Risk Premia based on Cross-Sectional Model Performance

49 Pages Posted: 16 Sep 2020 Last revised: 7 Jan 2021

Date Written: August 3, 2020

Abstract

This paper proposes a new parametric approach for estimating linear factor pricing models with time-varying risk premia. In contrast to most established methods, the framework presented abstains from specifying a time series model with external predictor variables. Instead, time-varying risk prices and exposures follow an observation-driven updating scheme that reduces the one-step-ahead prediction error from a cross-sectional factor model at the current observation. This agnostic approach is particularly useful in situations where predictors are unknown or of uncertain quality. Updating schemes for elliptically distributed returns are derived and propose cross-sectional regression errors as driving sequence for the parameter dynamics. Estimation and inference are performed by likelihood maximization. A simulation study confirms that the novel method is capable of filtering and predicting substantial risk price movements. The empirical performance is illustrated by an application to a macrofinance model of currency carry trades.

Keywords: Dynamic Asset Pricing, Generalized Auto-Regressive Score Models, Time- Varying Risk Premia, Return Predictability

JEL Classification: G12, G17, C58

Suggested Citation

Umlandt, Dennis, Score-Driven Asset Pricing: Predicting Time-Varying Risk Premia based on Cross-Sectional Model Performance (August 3, 2020). Available at SSRN: https://ssrn.com/abstract=3666324 or http://dx.doi.org/10.2139/ssrn.3666324

Dennis Umlandt (Contact Author)

University of Trier ( email )

15, Universitaetsring
Trier, 54286
Germany

HOME PAGE: http://risk.uni-trier.de

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

Paper statistics

Downloads
113
Abstract Views
581
rank
299,464
PlumX Metrics