Review of Quantitative Finance and Accounting, 13 (1999): 171±188
# 1999 Kluwer Academic Publishers, Boston. Manufactured in The Netherlands.
Random Walks and Market Ef®ciency Tests: Evidence
from Emerging Equity Markets
South Carolina State University, Orangeburg, SC 29117
Saint Louis University, St. Louis, MO 63108
JOHN A. COLE
Benedict College, Columbia, SC 29204
Abstract. We use the multiple variance-ratio test of Chow and Denning (1993) to examine the stochastic
properties of local currency- and US dollar-based equity returns in 15 emerging capital markets. The technique is
based on the Studentized Maximum Modulus distribution and provides a multiple statistical comparison of
variance-ratios, with control of the joint-test's size. We ®nd that the random walk model is consistent with the
dynamics of returns in most of the emerging markets analyzed, which contrasts many random walk test results
documented with the use of single variance-ratio techniques. Further, a runs test suggests that most of the
emerging markets are weak-form ef®cient. Overall, our results suggest that investors are unlikely to make
systematic nonzero pro®t by using past information in many of the examined markets, thus, investors should
predicate their investment strategies on the assumption of random walks. Additionally, our results suggest
exchange rate matters in returns' dynamics determination for some of the emerging equity markets we analyzed.
Key words: random walk, stock prices, multiple variance-ratio test, emerging capital markets, weak-form
JEL Classi®cation: G15, G14
The random walk properties of security prices have an important bearing on the
determination of security return dynamics and on associated potential trading strategies, as
is amply suggested by Poterba and Summers (1988, pp. 53±54), Lo and MacKinlay (1989),
and Eckbo and Liu (1993). Random walks, which are a special case of unit root processes,
help identify the kinds of shocks that drive stock prices. If a given equity price series is, for
instance, a random walk, the generating process is dominated by permanent components
and hence has no mean-reversion tendency.
A shock to the series from an initial
equilibrium will lead to increasing deviations from its long-run equilibrium. Moreover, the
random walk properties of stock returns are considered an outcome of the ef®cient market
hypothesis (i.e., stock prices exhibit unpredictable behavior, given available information).
Accordingly, Liu and Maddala (1992) demonstrate how the presence or absence of random
walks in security returns is crucial to both the formulation of rational expectation models
and the testing of market ef®ciency hypothesis.
Several studies, (e.g., Hakkio (1986), Summer (1986), Fama and French (1988), and
Poterba and Summers (1988)) demonstrate that standard random walk hypothesis