Conditional Beta Estimation and Forecasting with Panel Data Methods

40 Pages Posted: 7 Nov 2005

See all articles by Michelle L. Barnes

Michelle L. Barnes

Federal Reserve Bank of Boston

Anthony (Tony) W. Hughes

University of Adelaide

Date Written: April 2001

Abstract

Standard approaches to the estimation and testing of conditional CAPM models with time-varying or random beta have ignored the potential panel nature of financial data. We test for whether or not homogeneity restrictions on the time-variation component of multifactor betas and on the slope parameters for the conditioning variables can be rejected. We find that such homogeneity restrictions are not rejected, and show that there are resultant benefits for testing conditional CAPM and forecasting expected returns and beta. Further, this panel approach yields more precise parameter estimates, and a greater understanding of the significance of both conditional variables and multi-factors.

Keywords: conditional CAPM, panel data, heterogeneity, beta

JEL Classification: G12, C23, C52

Suggested Citation

Barnes, Michelle L. and Hughes, Anthony (Tony) W., Conditional Beta Estimation and Forecasting with Panel Data Methods (April 2001). Available at SSRN: https://ssrn.com/abstract=842645 or http://dx.doi.org/10.2139/ssrn.842645

Michelle L. Barnes (Contact Author)

Federal Reserve Bank of Boston ( email )

600 Atlantic Avenue
Boston, MA 02210
United States

Anthony (Tony) W. Hughes

University of Adelaide

233 North Terrace
Adelaide, South Australia
AUSTRALIA

Register to save articles to
your library

Register

Paper statistics

Downloads
387
Abstract Views
1,681
rank
76,278
PlumX Metrics