Random Walk in Thinly Traded Stock Markets: The Case of Kuwait

Abstract

A previous weak-form efficiency study that utilized conventional testing procedures found that the majority of Kuwaiti listed company shares appear to follow a random walk (Butler and Malalkah, 1992). Our study uses weekly market Index data and a combination of traditional and modern testing procedures to examine whether the previous results may be sample and/or methodology specific. The results of traditional tests were consistent with previous findings. However, when more robust econometric techniques were employed, the results consistently favored stationarity but not a random walk. Furthermore, the results indicate that the Kuwaiti stock price series are non-linear and can be described by a Generalised Autoregressive Conditional Heteroscedasticity (GARCH) Model.