SHORT AND LONG-TERM DYNAMICS OF HERD BEHAVIOUR AT THE JOHANNESBURG STOCK EXCHANGE
Olivier Niyitegeka, Regent Business School
Devi Datt Tewari, University of Zululand
This study uses the Autoregressive Distributed Lag (ARDL) approach to cointegration to examine the short- and long-term dynamics of investors’ herd behaviour at the JSE. The results from the ARDL model suggest that herding exists at the JSE. The study also noted that herd behaviour takes place with lapses in time; however the unrestricted error correction results suggest that such behaviour has a high speed of adjustment, implying that herding is a short-lived phenomenon. Since the direction of the market affects investors’ behaviour, the study also investigated the asymmetric effects of herding behaviour during rising versus falling markets. While herding behaviour was identified during a rising market, the results did not support the presence of herd behaviour in a falling market.