A dissertation submitted to the Faculty of Science, University of the Witwatersrand, Johannesburg, in fulfilment of requirements for the degree of Master of Science. January 2015. / Stylized facts play a significant role in the testing whether models agree with known statistical anomalies and phenomena that occur in financial markets or not. Thus, we can use these stylized facts as a modelling tool or just to understand the general behavior of financial markets better. In the paper by Bouchaud et al in 2004 [1] we see the promotion of a new stylized fact that correlations in trade signs fail to die out, even after large lags. In fact, Bouchaud et al expressed the correlations as a slow power-law decay over trade ticks. In the results of our empirical study of JSE and BM&FBOVESP we find that the selected stocks show the this same power-law decay of correlations of trade signs. We also find that the stocks behave in a way which may allow for price manipulation at high enough trading rates as discussed by Gatheral [2].

A research report submitted to the Faculty of Commerce, Law and Management, University of the Witwatersrand, in partial fulfilment of the requirements for the degree of Master of Finance and investment. Johannesburg, 2015 / A pairs trading strategy is a market neutral trading strategy that tries to make a profit by making use of inefficiencies in financial markets. In the equity pairs trading context, a market neutral strategy, is a strategy that hedges against both market and sector risk. According to the efficient market theory in its weak form, a pairs trading strategy should not produce positive returns since the actual stock price is reflected in its past trading data. The main objective of this paper is to examine the performance and risk of an equity pairs trading strategy in an emerging market context using daily, weekly and monthly prices on the Johannesburg Securities Exchange over the period 1994 to 2014. A bootstrap method is used determine whether returns from the strategy can be attributed to skill rather than luck. / MT2016

Thesis (M.M.(Finance & Investment)--University of the Witwatersrand, Faculty of Commerce, Law and Management, Wits Business School, 2016 / Co-location on the JSE took place on the 14th of May 2014. This dissertation looks at the impact this event has had on the market. In order to measure the effects of colocation, market quality factors are examined before and after the event to see whether there were any significant changes. A regression is then undertaken to see the correlation between co-location, liquidity and volatility. Our results suggest that colocation benefits market liquidity but we are unable to assess the relationship with volatility. This means that the growing liquidity in the market can be used to attract more institutions and firms wishing to run trading algorithms and strategies. Trades originally meant for dark pools can be now traded on the JSE co-location servers. By moving trades from dark pools to co-location servers at the JSE and encouraging institutions to use these facilities, transparency can be increased. Exchanges should implement kill switches if it is apparent that they are being impaired or flooded with erroneous orders. The deployment of kill switches, circuit breakers and other system compliance will improve investor confidence and market stability. Subsequent research can lead to better understanding by investigating the correlation between colocation and volatility. / MT 2018

Thesis (M.M. (Finance & Investment)--University of the Witwatersrand, Faculty of Commerce, Law and Management, Wits Business School, 2017 / The South African economy is built on the richness of mineral resources found in most parts of the country. In 2013, Chamber of Mines reported that the country earned about R 2.4 trillion from the export market over the past 10 years. However, the industry has also shown signs of financial ill health in recent years. This study examines the factors affecting the financial performance (return on capital invested, return on asset and stock price return) of mining companies in South Africa with a particular focus on employee related factors (number of employees, wage bill and safest statistics). The study examines 24 publicly listed companies over a 6 year period using panel data analysis. The results show that lost time injury rate, number of fatalities are significant variables in explaining the changes in financial performance. Labour indicators such as number of employees, lost time injury rate and wages have a negative relationship with all financial indicators. / GR2018

A research report submitted to the Faculty of Commerce, Law and Management, University of the Witwatersrand, Johannesburg, In partial fulfilment of the requirements for the degree of Master of Management (Finance and Investment Management), 2016 / The relationship between real economic activity and stock market performance is one that has been extensively researched throughout many decades, across many economies. Many issues and debates have stemmed involving this relationship, with the major ones including those of the significance of the relationship, nature of the relationship as well as causality and direction of causality within the relationship. This research paper examines this relationship within the South African context, comparing the pre and post 2008 global financial crisis periods. Results both in support of and contrary to theory were found as real economic activity had an immediate postitive response to shocks imposed on the stock index, whilst the stock index had an immediate negative response to shocks imposed on real economic activity. Through the use of granger causality testing, no causality was found in either direction. Furthermore, no major differences were noted between the pre and post crisis periods. / GR2018

A research report submitted to the Faculty of Commerce, Law and Management, University of the Witwatersrand in partial fulfilment of the requirements for the degree of Master of Commerce in Finance. Johannesburg, South Africa March 2016 / The study examined whether: share repurchase events and changes in dividends were good predictors of future changes in earnings. The research also investigated how the South African market reacted to share repurchase events in the short-run. Using INET BFA, data for 226 dividend paying companies and 55 share repurchasing companies, trading on the JSE during the period 2003 to 2013, was collected. Dividend theory suggests that changes in dividends convey information content about the future earnings of the firm. After testing this theory, limited support was found for this notion. Firms that had increased dividends at (T0) showed significant earnings increases in that year. Nonetheless, some of the dividend increasing firms showed no subsequent unexpected earnings growth at (T1) and (T2). While the size of the dividend increase had a strong positive relationship with current earnings; it failed to predict future earnings with any consistency. Firms that had cut dividends at (T0) experienced a reduction in earnings in that year but showed increases in earnings at (T1). However, consistent with Lintner‘s (1956) model on dividend policy, firms that had increased their dividends were less likely to experience a reduction in earnings, as opposed to the no-change or dividend decrease groups. A linear regression model was employed in testing whether share repurchases were useful in predicting changes in future earnings. According to the results reported in the regression model, share repurchases are a good predictor of future changes in earnings. The study at hand then went on to explore how the South African market reacted to share repurchases. Through the utilisation of the Market Model-Event Study Methodology (with an event window of 41 days, 20 days prior and 20 days post the event), the findings of the report indicated that the South African market reacted positively to share repurchases. This was evidenced through positive: share price returns, abnormal returns and average abnormal returns, post the event. Nonetheless, cumulative average abnormal returns remained negative in the short-run. In addition, the results showed that firms engage in share repurchase activities in order to signal that the stock is undervalued. There was an observable trend of declining share prices before the share repurchase event. A few recommendations were proposed following the results obtained. Dividends are unable to predict changes in earnings. Therefore, a dividend cut, is not an indication that a company‘s earnings will decrease in the future or that the managers of that company foresee a decline in future earnings. From a share repurchase point of view, managers of JSE listed companies should not only focus on the short-term benefits of share repurchase events. These benefits are generally short lived as shares do return to their falling state, however authors such as Wesson, Muller and Ward (2014) have shown that the benefits of share repurchase events can also be observed in the long- run, A further point to note for both investors and managers of JSE listed companies is that share repurchases are a good predictor of future earnings. Therefore, it is very confusing for investors when a company announces a share repurchase event but does not follow through with it. / MT2017

A dissertation submitted in fulfillment of the requirements for the degree in Masters in Management Finance and Investment , University of the Witwatersrand, Johannesburg, 2017 / This thesis examines the performance of value and growth stocks during the ten year period June 2006 to 2016 within five emerging markets countries namely South Africa, Nigeria, Brazil, India and Argentina. Value stocks are those stocks that trade at low prices in comparison to its fundaments value of the company and growth stocks are those stocks that trade at high prices compared to the company’s fundaments. The portfolios of value and growth stocks are created in the five abovementioned countries. The performance of value and growth stocks are studied by constructing portfolios on the basis of price-to-earnings, price-to-book, price-to-cash flow and price-earnings-growth. The data to calculate these price-multiples are derived from the audited statement of comprehensive income, statement of financial position and statement of cash flow of the companies. Trade data on listed stock, listed indices, cash dividends and risk-free rates are derived from mainly from Bloomberg.com and Morningstar.com. To classify stocks to be included in value or growth portfolios, a 30 percent cut-off is used. The portfolio returns and risk, price-multiples are studied as well to research whether one price-multiple provide higher return than others. Total return and risk-adjusted measures are studied by means of average daily returns to scrutinize which class of stocks, value or growth, provided the highest return. A regression analysis is performed to study if the Capital Asset Pricing model and a two-factor model can elaborate on the excess returns yield by value and growth portfolios. The findings are that value stock portfolio provide a higher total return than growth stocks portfolio. The value stocks as compared to growth stocks, also provide a fraction of higher return per unit of risk, as measured by Jensen’s Alpha and Treynor. The study also shows that value portfolios classified on price-to-book yield higher returns than portfolios constructed on other price multipliers. The regression analyses show that the CAPM two-factor model is able to explain the excess returns on value and growth portfolios. The beta coefficients of value stocks are higher than growth stocks, which is consistent with the general theory that higher betas found in stocks should, by definition, produce higher returns, this also suggest that the reason behind the of outperformance by value stocks over growth stocks is a compensation of risk. While value and growth stocks are studied over a period of 10 years on five emerging markets there is some limitations and implications for future research exist. One major limitation concern is the sample size of 5 emerging markets out of 152 emerging and developing countries as listed by the International Monetary Fund. Therefore reaching statistical conclusion makes it difficult to generalize towards other countries. / MT 2018

This three year indepth study was prompted after a decade of working as a corporate advisor for numerous stockbroking firms' corporate advisory and listing divisions. An overwhelming lack of discernible pricing methodology for IPOs on the JSE's Main Board and failed Venture Capital and Development Capital Markets was transferred to the new Alternative Exchange (AltX). This prompted lengthly discussions with former head of JSE's AltX Noah Greenhill. Such discussions are set out in this dissertation and relate to pricing methodologies and the lack of guidance or legislation as set out in the JSE's schedule 21 of Listing requirements. The focus of this dissertation is thus centred on whether the current adopted methodologies to establish a fair and reasonable pre-IPO share price is effective. To achieve this, global pricing methodologies were assessed within the framework of various valuation techniques used by South African Designated Advisors.

Thesis (MBA)--Stellenbosch University, 1999. / ENGLISH ABSTRACT: This is a comparative study to determine the accuracy of the cumulative factors calculated and used by INET and the University of Stellenbosch, Graduate School of Business (USB). These factors are calculated whilst taking into account the changes in capital structure due to the effects of share splits (splits), consolidations and capitalisation issues in the calculation of dividends per share (both interim and final), and closing share prices. For this purpose the data of 350 listed industrial companies on the Johannesburg Stock Exchange was evaluated over a 28-year period (1970-1997). In cases where a company was delisted before the date of the financial year-end 1997, that company was deleted from the study for the full period. The analysis of share splits, consolidations and capitalisation issues in the calculation of a cumulative factor for the determination of dividends per share and closing share prices, is therefore of critical importance. It is important to the companies as well as parties who are involved in maintaining data of listed companies on the Johannesburg Stock Exchange. It is also important to those who use this data for research purposes. South African studies using dividends per share and share prices from INET or the USB assume that the data is accurate. This study is an effort to verify the accuracy of the two mentioned databases. The results of the study clearly suggest or indicate that there are indeed numerous inaccuracies (differences) between the data kept by both INET and USB databases. It is therefore important that the data be revisited so that these anomalies can be rectified. / AFRIKAANSE OPSOMMING: Hierdie is 'n vergelykende studie om die akkuraatheid te bepaal van die kumulatiewe faktore soos deur INET en die Universiteit van Stellenbosch se Nagraadse Bestuurskool (USB) bereken en gebruik. Hierdie faktore word bereken om die effek van die onderverdeling en konsolidasie van aandele, asook kapitalisasie-uitgifte op die dividend per aandeel (beide interim en finaal) en sluitingsaandeelpryse te bepaal. Vir hierdie doel was die data van 350 industriele maatskappye wat op die Johannesburgse Aandelebeurs genoteer is oor 'n 28-jaar periode (1970-1997) geevalueer. In gevalle waar die maatskappye voor die finansiele jaareinde 1997 gedenoteer is, is die maatskappy uit die studie weggelaat. Die analise van onderverdeling en konsolidasie van aandele en kapitalisasie-uitgifte in die berekening van 'n kumulatiewe faktor vir die bepaling van dividend per aandeel en die sluitingsaandeelpryse, is van kritiese belang. Dit is belangrik vir die maatskappye en ander belanghebbendes wat gemoeid is met die instandhouding van data van genoteerde maatskappye op die Johannesburgse Aandelebeurs. Dit is ook van belang vir diegene wat die data vir navorsingsdoeleindes gebruik. Suid-Afrikaanse studies wat op dividend per aandeel en aandeelpryse van INET of die USB gebaseer is, veronderstel dat die data korrek is. Hierdie studie poog om die akkuraatheid van die genoemde twee databasisse te verifieer. Die resultate van die studie toon duidelik aan dat daar 'n hele aantal onakkuraathede (verskille) tussen die data onderhou deur beide INET en die USB bestaan. Dit is dus belangrik dat die data weer ondersoek word ten einde verskille uit die weg te ruim.

Thesis (MBA)--Stellenbosch University, 2000. / Dividends are probable the most controversial subject in the finance literature. Dividends are paid cent for cent from company profits. Besides having tax implications on the company, they reduce sources of internal finance for the company. On the other hand, the value of a company is the net present value of cash flows. Theoretically a company that does not pay dividends now and at anytime in the future has a value of zero. Companies have dividend policies applicable to themselves. It is therefore valid to argue that the revision of the dividend policy has an underlying reason. This study was done to investigate the effect of unexpected dividend policy changes to daily share price movement. This study seeks to establish the validity of the much-debated subject of information significance of dividends. Past studies at most failed to converge to an agreement on information significance of dividends. The investigation revealed that there were no significant abnórmal returns earned on the announcement date on three out of four instances. However the results of the cumulative abnormal returns revealed that share prices react to dividend changes during the period of investigation i.e. twenty days before and twenty days after the announcement. The overall adjustment in share prices over the period studied is in the same direction as the dividend charge. The investigation also revealed that price adjustments take place before and after the announcement date. Price adjustments on the Johannesburg Stock Exchange were not efficient as compared to adjustments on the New York Stock Exchange. Share price adjustments on the New York Stock Exchange took place during the period of a day before and a day after the announcement. The lack of similarity can be attributed to either sophistication of the market participants or efficiency of the Johannesburg Stock Exchange.