Statistics and Its Interface

Volume 1 (2008)

Number 2

Quantile momentum

Pages: 243 – 254

DOI: https://dx.doi.org/10.4310/SII.2008.v1.n2.a3

Authors

Gilbert W. Bassett (Department of Finance, University of Illinois at Chicago, Chicago, Il., U.S.A.)

Rong Chen (Department of Statistics, Rutgers University, New Brunswick, New Jersey, U.S.A.)

Yongchang Feng (Department of Business Statistics and Econometrics, Peking University, Beijing, China)

Abstract

A stock portfolio based on momentum strategies buys stocks that have recently performed well and sells (or shorts) stocks that have recently performed poorly. The most commonly used measure of past performance (momentum) of a stock is its average return over the previous 2 to 12 months. In this paper, we propose a different set of measures of past performance based on the quantiles of past returns and investigate the performance of momentum portfolios based on such measures. We also introduce a robust version of the proposed quantile momentum by locally smoothing data before ranking. It is shown that the portfolios under these proposed alternative momentum measures can have very different returns from one another, as well as from the standard portfolio based on the average. We also consider the correlations of portfolio returns between the alternative strategies and their combinations. A well known feature of momentum portfolios is that they carry incidental β exposures depending on whether the market has been rising or falling. A practical matter for portfolio managers is the extent to which momentum can be improved by neutralizing the incidental β exposures. We investigate how the various definitions of momentum are likely to affect the incidental β exposures in long and short momentum portfolios.

Keywords

momentum strategies, betasensitive portfolio, Sharpe ratio, quantile momentum, price inefficiency

Published 1 January 2008