| Summary: | This study revisits the efficient market hypothesis (EMH) with regard to the Kuala Lumpur Stock Exchange (KLSE) at the sectoral level. Based on Liu and Narayan's (2011) GARCH-based unit-root with structural breaks test, the unit-root null is rejected for all except one sector. By contrast, models based on commonly used unit-root tests that ignore heteroskedastic and/or breaks tend to favour the EMH. We find that the half-life estimates based on the local-persistent model are short, with the majority of them taking less than six months to absorb half a shock. All in all, the indices examined are largely inconsistent with weak-form efficiency, which implies that the returns on equity portfolios are indeed predictable.
|
|---|
