“Predictable Stock Returns: The Role of Small Sample Bias”, Journal of Finance,
INTRODUCTION
The
study focuses on predictable sock returns; the role of small sample bias. Small
sample bias in asymptotic standard errors in the context of overlapping
observations on multiperiod returns has received attention in several recent
papers which conclude that they are too small; see Hodrick (1991), Kim, Nelson,
and Startz (1991), Richardson and Smith (1991), and Richardson and Stock
(1989).
However, the potential for small sample bias in the regression coefficient has not received corresponding attention. The existence of small sample bias in tests of predictability has been pointed out by Mankiw and Shapiro (1986) and Stambaugh (1986) who showed that regression on predetermined variables will reject the null hypothesis of nonpredictability too often. It is in line with the above that this study sought to evaluate the possibility that small sample bias could be playing an important role in the inference that stock returns are predictable from fundamentals and in estimates of the degree of predictability.
However, the potential for small sample bias in the regression coefficient has not received corresponding attention. The existence of small sample bias in tests of predictability has been pointed out by Mankiw and Shapiro (1986) and Stambaugh (1986) who showed that regression on predetermined variables will reject the null hypothesis of nonpredictability too often. It is in line with the above that this study sought to evaluate the possibility that small sample bias could be playing an important role in the inference that stock returns are predictable from fundamentals and in estimates of the degree of predictability.
To get further details on the topic...leave your comment or follow me on twitter @anacom_okey
Subscribe to:
Post Comments
(
Atom
)
No comments :
Post a Comment