Abstract

 
 

References (65)



 
 

Citations (2)



 


 



Can An Investor Profit from Return Predictability in Real Time?


Yufeng Han


University of Colorado at Denver - Business School

March 2005


Abstract:     
Despite the extensive empirical evidence on return predictability, the question whether investors can profit from return predictability in real time remains unsettled. One problem is that the predictive model is likely misspecified, which has received little attention. By taking a more careful approach that guards against model mis-specification, we study the economic significance of return predictability by evaluating extensively the out-of-sample portfolio performance of various predictive models and variations of predictive variables. We find that investors can benefit from predicting the sign, but not the magnitude, of the market expected returns only in periods when the market performs poorly. We also find that there is no significant performance difference among the predictive models, although the best statistical model often performs the worst and a GARCH feature seems important.

Number of Pages in PDF File: 36

Keywords: Return predictability, Economic value, Portfolio performance, nonlinearity

working papers series


Download This Paper

Date posted: December 1, 2004  

Suggested Citation

Han, Yufeng, Can An Investor Profit from Return Predictability in Real Time? (March 2005). Available at SSRN: http://ssrn.com/abstract=625344 or http://dx.doi.org/10.2139/ssrn.625344

Contact Information

Yufeng Han (Contact Author)
University of Colorado at Denver - Business School ( email )
1475 Lawrence St.
Denver, CO 80204
United States
303-3158458 (Phone)
Feedback to SSRN (Beta)


Paper statistics
Abstract Views: 711
Downloads: 205
Download Rank: 72,963
References:  65
Citations:  2

© 2013 Social Science Electronic Publishing, Inc. All Rights Reserved.  FAQ   Terms of Use   Privacy Policy   Copyright
This page was processed by apollo6 in 1.453 seconds