Project (M.B.A.) - Simon Fraser University, 2004. / Theses (Faculty of Business Administration) / Simon Fraser University. MBA-GAWM Program. Senior supervisor: Dr. Robert R. Grauer.

Magister Commercii - MCom / Since the discovery of herding behaviour in financial markets in the 1990s, it has become an area of interest for many investors, practitioners and scholars. Herding behaviour occurs when investors and market participants trade in the same direction during the same time period, as a result of the influence of other investors. Studies on herding behaviour have been undertaken in both the developed and developing economies and majority of these studies have confirmed the existence of herding behaviour in the stock markets. Despite its tremendous growth, the South African financial markets are not immune to such market anomaly. Herding behaviour on the JSE was first investigated in 2002 focusing in the unit trust industry on the South African stock market. Motivated by this, this study assessed the presence of herding behaviour using the Johannesburg Securities Exchange tradable sector indices. Four indices were employed, namely Financials, Industrials and Resources and were benchmarked against the JSE All Share Index for the period from January 2007 to December 2017. The industrials index ((FINI15) constitutes of 25 largest industrial stocks by market capitalization, the financials index (FINI15) comprises of 15 largest financial stocks by market capitalization, the resources index (RESI10) which represents 10 largest resources stocks by market capitalization and lastly the FTSE/JSE All Share Index defined as a market capitalization-weighted index which is made up of 150 JSE listed companies and is the largest index in terms of size and overall value JSE. The FTSE/JSE All Share Index was used as a benchmark for investors to check how volatile an investment is. The South African economy experienced the effects of the 2008 global financial crisis from 01 July 2007 to 31 August 2009. This study split the examination period into three categories namely before the global financial crises which was the period starting from 1 January 2007 to 30 June 2007, then the period during the global financial crisis which was from 1 July 2007 to 31 August 2009 and lastly the period after the global financial crises which was from 1 September 2009 to 31 December 2017. Apart from the diversity of the indices, the length of the examination period also had a significant influence towards the magnitude of herding behaviour on the JSE.

M.Com. (Financial Management) / This paper studies the effects of bull and bear market states on the profitability of a momentum investment strategy. That is, a strategy that buys past winners and sells past losers is simulated over the period 3 July 2002 to 8 August 2012 and its profitability is reviewed in light of bull and bear sub-periods. Such an investment strategy has been shown to yield abnormal returns in several markets around the world, including the South African stock market. By doing so, these studies challenge the efficient market hypothesis, a central and widely accepted hypothesis within traditional portfolio theory. There are many theories that have been used to explain why abnormal profits are achievable using a momentum investment strategy. By determining the effects of bull and bear market states on the profitability of a momentum investment strategy, this paper provides some insight into which theories, if any, are most relevant to the South African stock market context. It is found that on average, a momentum portfolio yields abnormal returns over the full sample period, with the chief driver of these returns being the winner component of the portfolio. When broken into bull and bear sub-periods, it was found that a momentum investment strategy only yields abnormal returns during a bull period, whilst these abnormal returns became negative during a bear period. These results are consistent with one efficient market hypothesis explanation and two behavioral models presented in past studies. The results indicate that the market may be efficient and that changes in macroeconomic risk are the cause of momentum profits. However, insofar as the macroeconomic risk explanation is inaccurate, these results support the behavioural models of Daniel, Hirshleifer, and Subrahmanyam (1998); and Hong and Stein (1999). Both these models predict that momentum returns will be strongest during bull periods.