Abstract
Purpose - The purpose of this paper is to re- examine the presence of rational speculative bubbles in the Malaysian stock market in light of contradictory results presented in previous studies.
Design/methodology/approach - The authors use descriptive statistics, explosiveness tests and the duration dependence test. They use an expanded data set that encompasses at least two alleged bubble episodes addressing a significant limitation of previous studies. The authors use both monthly and weekly returns addressing concerns about the sensitivity of duration dependence test results to the use of monthly versus weekly returns, as well as a battery of alternative measures of returns.
Findings - The authors detect bubble footprints but they do not appear to be rational. They found no evidence of rational speculative bubbles over the sample period regardless of whether monthly or weekly returns was used. The authors suggest that if there were bubbles in the Malaysian stock market, they might have been caused by irrational investor behaviour. The authors' results do not support the suggestion that the duration dependence test is sensitive to the use of monthly versus weekly returns.
Practical implications - Despite the absence of rational bubbles in the Malaysian stock market, the faint bubble footprints detected still suggest caution for investors, as the authors cannot categorically rule out the presence of irrational bubbles.
Originality/value - This paper clarifies conflicting results of previous studies. It also contributes to the literature on bubble testing by presenting new evidence from an emerging market refuting the claim that duration dependence test results are sensitive to the use of either weekly or monthly returns.