Thesis (MBA) -- Stellenbosch University, 2007. / ENGLISH ABSTRACT: Hedge funds are amongst the fastest growing types of investment funds, both in tenns of worldwide assets under management, as well as the number of private and institutional investors. More recently, analysts and investors focussed their attention on accurately estimating the inherent risks of hedge funds (e.g, Brooks & Kat, 2001; Fung & Hsieh, 2004). Past research suggests that the traditional approach of assessing the risks of investment funds through mean-variance analysis can lead to severe underestimation of left-hand-tail risks for hedge funds (Amenc, Malaise, Martellini & Vaissie, 2004; Favre & Galeano, 2002; Fung & Hsieh, 1999). This phenomenon is mainly attributab le to the non-normal distribution of monthly hedge fund returns around the mean. In addition, it has been found that skewed return distribution with high excess kurtosis has substantial impact on the rel iability of beta as a measure of systemic risk in hedge funds (Chan, Getmansky, Haas & Lo, 2005). Other problems when estimating hedge fund risks arise from serial correlation of time series (Getmansky, Lo & Makarov, 2003), managerial and survivorship bias (Amin & Kat, 2001 ), as well as spurious bias when estimating performance from economic time series (Fung & Hsieh, 2000). The following thesis provides statistical evidence of the limitations of traditional risk measures when applied to hedge fund investments. It also includes advice on how to improve the significance of the aforementioned risk measures. In the course of the mean-variance analysis, the applicability and reliability of Value at Risk as a risk measurement tool for hedge funds is explored. Furthennore, the reliability and accuracy of different univariate and multivariate regression models is tested. In the final chapter emphasis is placed on the possibilities of predicting the inherent risks of single funds from hedge fund style index performance. This should provide investors and analysts with an introductory framework for the appropriate risk assessment of hedge funds, considering the unique structure and dynamics of these alternative investment funds. / AFRIKAANSE OPSOMMING: Skansfondse tel onder die vinnigste groeiende tipes beleggingsfondse in terme van sowel wereldwye bates onder bestuur as die aantal private en institusionele beleggers. OnJangs het analiste en beleggers hulle aandag daarop begin toespits om die inherente risiko's verbonde aan skansfondse akkuraat te bereken (Brooks & Kat. 2001; Fung & Hsieh, 2004). Vroeere navorsing het daarop gedui dat die tradisioncle benadering om die risiko's verbonde aan beleggingsfondse deur gemiddeldevariansie-analise te takseer, daartoe kan lei dat linkerkantse-eindrisiko's verbonde aan skansfondse emstig onderskat word (Fung & Hsieh, 1999; Favre & Galeano, 2002; Amenc. Malaise, Martellini & Vaissie, 2004). Hierdie verskynsel is hoofsaaklik toe te skryf aan die abnonnale verspreiding van maandeliksc skansfondsopbrengste rondom die gemiddelde. Boonop is bevind dat skewe verdeling met hoe kurtose-oorskryding aansienlik inslaan op die betroubaarheid van beta as 'n meting van sistemiese risiko by skansfondse (Chan. Getmansky. Haas & Lo, 2005). Ander probleme by die raming van skansfondsrisiko's spruit uit tydreekskorrelasie (Getmansky, Lo & Markov, 2003), bestuurs- en oorlewingsydigheid (Amin & Kat, 2002) en vals sydigheid by die beraming van prestasie uil die ekonomiese tydsreeks (Fung & Hsieh, 2000). Hierdie tesis gaan statistiese bewyse lewer van die tradisioncle risikometings se beperkings wanneer dit op skansfondsbeleggings toegepas word. Verder sal daar raad gegee word oor hoe om die beduidendheid van die genoemde risikometings te verbeter. In die loop van die gemiddeldevariansie-analise sal die toepasbaarheid en betroubaarheid van die Waarde onder Risiko as 'n risikometing vir skansfondse ondersoek word. Voorts sal die betroubaarheid en akkuraatheid van verskillende ecnvariaat- en meervariaatregressiemodelle getoets word. In die laaste hoofstuk val die klem op die moontlikheid om die inherente risiko's van enkelfondse aan die hand van 'n skansfondstipe-indeksprestasie te voorspel. Wat hier volg, behoort beJeggers en analistc van 'n inleidende raamwerk vir die toepaslike risikotaksering van skansfondse - met inagneming van die unieke struktuur en dinamika van hierdie altcmatiewe beleggingsfondse - te voorsien.

Ph.D. / This thesis investigates the persistence of hedge fund managers’ skills, the optimality of strategies they use to outperform consistently the market during periods of boom and/or recession, and the market risk encountered thereby. We consider a data set of monthly investment strategy indices published by Hedge Fund Research group. The data set spans from January 1995 to June 2010. We divide this sample period into four overlapping sub- sample periods that contain different economic market trends. We define a skilled manager as a manager who can outperform the market consistently during two consecutive sub-sample periods. To investigate the presence of managerial skills among hedge fund managers we first distinguish between outperformance, selectivity and market timing skills. We thereafter employ three different econometric models: frequentist, Bayesian and fuzzy regression, in order to estimate outperformance, selectivity and market timing skills using both linear and quadratic CAPM. Persistence in performance is carried out in three different fashions: contingence table, chi-square test and cross-sectional auto-regression technique. The results obtained with the first two probabilistic methods (frequentist and Bayesian) show that fund managers have skills to outperform the market during the period of positive economic growth (i.e. between sub-sample period 1 and sub-sample period 3). This market outperformance is due to both selectivity skill (during sub-sample period 2 and sub-sample period 3), and market timing skill (during sub-sample period 1 and sub- sample period 2). These results contradict the EMH and suggest that the “market is not always efficient,” it is possible to make abnormal rate of returns.However, the results obtained with the uncertainty fuzzy credibility method show that dispite the presence of few fund managers who possess selectivity skills during bull market period (sub-sample period 2 and sub-sample period 3), and market timing skills during recovery period (sub-sample period 3 and sub-sample period 4); there is no evidence of overall market outperformance during the entire sample period. Therefore the fuzzy credibility results support the appeal of the EMH according to which no economic agent can make risk-adjusted abnormal rate of return. The difference in findings obtained with the probabilistic method (frequentist and Bayesian) and uncertainty method (fuzzy credibility theory) is primarily due to the way uncertainty is modelled in the hedge fund universe in particular and in financial markets in general. Probability differs fundamentally from uncertainty: probability assumes that the total number of states of economy is known, whereas uncertainty assumes that the total number of states of economy is unknown. Furthermore, probabilistic methods rely on the assumption that asset returns are normally distributed and that transaction costs are negligible.

Thesis submitted in fulfilment of the requirements for the degree of Master of Management in Finance and Investments in the FACULTY OF COMMERCE, LAW AND MANAGEMENT WITS BUSINESS SCHOOL at the UNIVERSITY OF THE WITWATERSRAND / The South African hedge fund industry is reported to have had R52 billion (USD 4.8 billion) assets under management at the end of December 2013. This compares to the global industry which is reported to have surpassed USD 2.6 trillion at the end of 2013. Due to the relative infancy of the local industry, little research exists to analyse the performance of South African hedge fund strategies. This study focuses on the performance of South African hedge fund strategies under different market regimes, taking into consideration market and economic factors specific to South Africa. The analysis shows that the hedge fund strategies offer a diversification benefit to more traditional asset classes, and the results of the study can be used to inform an investor’s allocation decision. The findings of the analysis are used as the basis of a portfolio construction framework for constructing a portfolio of hedge funds. The framework is predicated on the investor having a view on the forthcoming macro environment. The framework enables the investor to identify funds and strategies that have produced a stable alpha over a similar market regime for inclusion in the portfolio of funds. After identifying those funds and strategies most suited to the anticipated macro environment, the number of funds to be included in the portfolio is taken under consideration to determine the optimal number such that the performance and risk characteristics of the portfolio are not compromised. The analysis takes the higher moments of the distribution into account to cater for the non-normal nature of hedge fund distributions.

This dissertation investigates two fundamental questions related to how well exchange-traded funds that hold portfolios of fixed-income assets (bond ETFs) proxy for their underlying portfolios. The first question involves price/net-asset-value (NAV) mean-reversion asymmetries and the effectiveness of the arbitrage mechanism of bond ETFs. Methodologically, to answer the first question I focus on a time-series analysis. The second question involves the degree to which average returns of bond ETF shares respond to changes in factors that have been found to drive average returns of bond portfolios. To answer this question I shift the focus of the analysis to a cross-section asset pricing test. In other words, do bond ETF share prices track the value of their underlying assets, and are they priced by investors like bonds in the cross-section? The first essay concludes that bond ETF shares exhibit mean-reversion asymmetries when price and NAV diverge, along persistent small premiums. These premiums appear to reflect the added value that bond ETFs bring to the fixed-income asset market through smaller trading increments, greater liquidity, and the ability to buy on margin and sell short. The second essay concludes that market, bond-specific, and firm-specific risk factors can help to explain the variation in U.S. bond ETF average returns, but only size seems to be priced in the cross-section of expected returns. This is not surprising as the sample used in the asset pricing tests is limited to the period 2007-2010, which corresponds to the "great recession", and size has been interpreted in the asset pricing literature as a state variable that proxies for financial distress and is highly dependent on the phase of the real business cycle. / The two essays together suggest that bond ETFs can be used in trading strategies based on taking long and short positions in fixed-income assets, especially when trading in portfolios of fixed-income assets directly is not feasible. / by Charles W. Evans. / Thesis (Ph.D.)--Florida Atlantic University, 2011. / Includes bibliography. / Electronic reproduction. Boca Raton, Fla., 2011. Mode of access: World Wide Web.

Some commentators attribute the success of certain hedge fund activism events to “wolf pack” activism, the support offered by other investors, many of whom are thought to accumulate stakes in the target firms before the activists’ campaigns are publicly disclosed. This paper investigates wolf-pack activism by considering the following questions: Is there any evidence of wolf-pack formation? Is the wolf pack formed intentionally (by the lead activist) or does it result from independent activity by other investors? Does the presence of a wolf pack improve the activist’s ability to achieve its stated objectives? First, I find that investors other than the lead activist do in fact accumulate significant share-holdings before public disclosure of activists’ campaigns, a result consistent with wolf-pack formation. Second, these share accumulations are more likely to be mustered by the lead activist rather than occurring spontaneously. Notably, for example, the other investors are more likely to be those who had a prior trading relationship with the lead activist. Third, the presence of a wolf pack is associated with a greater likelihood that the activist will achieve its stated objectives (e.g., will obtain board seats) and higher future stock returns over the duration of the campaign.

This thesis examines the degree to which alternative funds deviate from their style-benchmark and how this is related to past performance and fund size, and how it impacts future risk and returns. Additionally the thesis examines how security selection and market timing skills differ across varying degrees of deviation from the benchmark. The thesis uses data for hedge funds, funds of funds, and CTA funds from the Center for International Securities and Derivatives Markets and employs fund???s tracking error relative to their style-benchmark to estimate the level of drift. The style-benchmarks used are the median return for all reporting funds that follow a particular style and funds are assigned a benchmark based on their self-reported style. First, this thesis documents statistically significant differences in the tracking errors of portfolios of funds with the highest tracking error versus funds with the lowest tracking error, implying that some managers drift from their self-reported style-benchmarks. Second, funds??? benchmark-inconsistency is less severe in the case of funds that have a regulatory obligation to disclose their performance, suggesting that the absence of regulation fosters an environment where managers can be more flexible with their investment approach. Third, the tendency to drift from the benchmark is most prevalent amongst funds with superior past performance as well as small funds. Fourth, future total portfolio risk increases as funds display more benchmarkinconsistency, suggesting that managers adopt riskier strategies as they attempt to enhance returns. Fifth, the thesis demonstrates that CTA funds that display drift from their benchmark produce higher absolute and relative returns in subsequent periods regardless of the direction of the general market. In contrast, the findings show for hedge funds and funds of funds, benchmark-inconsistent funds are likely to outperform in bull markets and underperform in bear markets. Finally, this thesis shows that more benchmark-consistent managers have better security selection skill. The main contribution of this thesis is in identifying the group of hedge funds, funds of funds, and CTA funds that are likely to deviate from their self-reported style-benchmark and the risk-return consequences of such deviations. The findings have implications for investors and regulators.

This thesis examines the degree to which alternative funds deviate from their style-benchmark and how this is related to past performance and fund size, and how it impacts future risk and returns. Additionally the thesis examines how security selection and market timing skills differ across varying degrees of deviation from the benchmark. The thesis uses data for hedge funds, funds of funds, and CTA funds from the Center for International Securities and Derivatives Markets and employs fund???s tracking error relative to their style-benchmark to estimate the level of drift. The style-benchmarks used are the median return for all reporting funds that follow a particular style and funds are assigned a benchmark based on their self-reported style. First, this thesis documents statistically significant differences in the tracking errors of portfolios of funds with the highest tracking error versus funds with the lowest tracking error, implying that some managers drift from their self-reported style-benchmarks. Second, funds??? benchmark-inconsistency is less severe in the case of funds that have a regulatory obligation to disclose their performance, suggesting that the absence of regulation fosters an environment where managers can be more flexible with their investment approach. Third, the tendency to drift from the benchmark is most prevalent amongst funds with superior past performance as well as small funds. Fourth, future total portfolio risk increases as funds display more benchmarkinconsistency, suggesting that managers adopt riskier strategies as they attempt to enhance returns. Fifth, the thesis demonstrates that CTA funds that display drift from their benchmark produce higher absolute and relative returns in subsequent periods regardless of the direction of the general market. In contrast, the findings show for hedge funds and funds of funds, benchmark-inconsistent funds are likely to outperform in bull markets and underperform in bear markets. Finally, this thesis shows that more benchmark-consistent managers have better security selection skill. The main contribution of this thesis is in identifying the group of hedge funds, funds of funds, and CTA funds that are likely to deviate from their self-reported style-benchmark and the risk-return consequences of such deviations. The findings have implications for investors and regulators.

This thesis examines the degree to which alternative funds deviate from their style-benchmark and how this is related to past performance and fund size, and how it impacts future risk and returns. Additionally the thesis examines how security selection and market timing skills differ across varying degrees of deviation from the benchmark. The thesis uses data for hedge funds, funds of funds, and CTA funds from the Center for International Securities and Derivatives Markets and employs fund???s tracking error relative to their style-benchmark to estimate the level of drift. The style-benchmarks used are the median return for all reporting funds that follow a particular style and funds are assigned a benchmark based on their self-reported style. First, this thesis documents statistically significant differences in the tracking errors of portfolios of funds with the highest tracking error versus funds with the lowest tracking error, implying that some managers drift from their self-reported style-benchmarks. Second, funds??? benchmark-inconsistency is less severe in the case of funds that have a regulatory obligation to disclose their performance, suggesting that the absence of regulation fosters an environment where managers can be more flexible with their investment approach. Third, the tendency to drift from the benchmark is most prevalent amongst funds with superior past performance as well as small funds. Fourth, future total portfolio risk increases as funds display more benchmarkinconsistency, suggesting that managers adopt riskier strategies as they attempt to enhance returns. Fifth, the thesis demonstrates that CTA funds that display drift from their benchmark produce higher absolute and relative returns in subsequent periods regardless of the direction of the general market. In contrast, the findings show for hedge funds and funds of funds, benchmark-inconsistent funds are likely to outperform in bull markets and underperform in bear markets. Finally, this thesis shows that more benchmark-consistent managers have better security selection skill. The main contribution of this thesis is in identifying the group of hedge funds, funds of funds, and CTA funds that are likely to deviate from their self-reported style-benchmark and the risk-return consequences of such deviations. The findings have implications for investors and regulators.

<p>Reading the daily financial news, it becomes quite obvious that hedge funds are receiving huge attention by financial analysts and politicians. Many people fear the influence of hedge funds on single companies as well as on the global economy.</p><p>This study does not judge the behavior of hedge funds. Instead, it focuses on the nature of hedge funds and their mystique image. Especially the common view of their market neutral performance is of interest, which is theoretically achieved through the use of derivatives and short positions. In this thesis the feature of market neutrality is investigated in depth, since it can improve the overall performance of investors’ portfolios in bull as well as in bear markets through diversification effects.</p><p>Therefore the hedge funds and their environment, the capital markets, are examined from an academically point of view by emphasizing on the following research questions:</p><p>1. Are hedge funds performing market neutral in bull and bear markets?</p><p>2. To what extent should they be included in optimal risky portfolios according to Modern Portfolio Theory and advanced performance measurement tools, considering their degree of market neutrality?</p><p>This study is based on extensive knowledge of financial and econometric theories. Capital market theories, modern portfolio theory, hedge fund data and econometric knowledge about time series analysis build the basis for further investigations and are necessary to understand the characteristics of hedge funds and hedge fund data.</p><p>In order to be able to deal with the shortcomings of hedge fund data, an analytical framework for the preparation of data is created that enables the authors to start with the analysis of these questions. The framework is applicable to all kinds of hedge funds presented in this thesis and enables the reader to test further hedge fund classes by himself.</p><p>In a quantitative study the created framework is applied to 2160 hedge funds of Barclays Hedge Fund Database, which builds the basis for analyzing market neutrality. Further input for the portfolio optimization consists of 19 hedge fund indices, which were provided by the Greenwich Alternative Investment Hedge Fund Database and 4 benchmark indices for the stock and bond market.</p><p>The analysis consists of two different parts. For the first research question various correlation and return matrices are constructed, which shall provide information about market neutrality of hedge funds. A correlation matrix also serves as important input for the portfolio analysis and therefore builds the basis for the analysis of the second research question. This shall provide some fundamental recommendations about the weighting of diverse hedge fund classes in optimal risky portfolios.</p><p>The conducted analysis demonstrated clearly the following findings:</p><p>1. Market neutrality has to be rejected for most hedge fund strategies. It is only attainable through strategies, which focus more on arbitrage and/or the bond market and therefore seems to be more a by-product than an actually provoked feature.</p><p>2. Only two strategies, equity short and convertible arbitrage, managed to beat the benchmark and to improve the overall performance of the portfolio when taking the specific return distribution of hedge funds into account.</p>

Reading the daily financial news, it becomes quite obvious that hedge funds are receiving huge attention by financial analysts and politicians. Many people fear the influence of hedge funds on single companies as well as on the global economy. This study does not judge the behavior of hedge funds. Instead, it focuses on the nature of hedge funds and their mystique image. Especially the common view of their market neutral performance is of interest, which is theoretically achieved through the use of derivatives and short positions. In this thesis the feature of market neutrality is investigated in depth, since it can improve the overall performance of investors’ portfolios in bull as well as in bear markets through diversification effects. Therefore the hedge funds and their environment, the capital markets, are examined from an academically point of view by emphasizing on the following research questions: 1. Are hedge funds performing market neutral in bull and bear markets? 2. To what extent should they be included in optimal risky portfolios according to Modern Portfolio Theory and advanced performance measurement tools, considering their degree of market neutrality? This study is based on extensive knowledge of financial and econometric theories. Capital market theories, modern portfolio theory, hedge fund data and econometric knowledge about time series analysis build the basis for further investigations and are necessary to understand the characteristics of hedge funds and hedge fund data. In order to be able to deal with the shortcomings of hedge fund data, an analytical framework for the preparation of data is created that enables the authors to start with the analysis of these questions. The framework is applicable to all kinds of hedge funds presented in this thesis and enables the reader to test further hedge fund classes by himself. In a quantitative study the created framework is applied to 2160 hedge funds of Barclays Hedge Fund Database, which builds the basis for analyzing market neutrality. Further input for the portfolio optimization consists of 19 hedge fund indices, which were provided by the Greenwich Alternative Investment Hedge Fund Database and 4 benchmark indices for the stock and bond market. The analysis consists of two different parts. For the first research question various correlation and return matrices are constructed, which shall provide information about market neutrality of hedge funds. A correlation matrix also serves as important input for the portfolio analysis and therefore builds the basis for the analysis of the second research question. This shall provide some fundamental recommendations about the weighting of diverse hedge fund classes in optimal risky portfolios. The conducted analysis demonstrated clearly the following findings: 1. Market neutrality has to be rejected for most hedge fund strategies. It is only attainable through strategies, which focus more on arbitrage and/or the bond market and therefore seems to be more a by-product than an actually provoked feature. 2. Only two strategies, equity short and convertible arbitrage, managed to beat the benchmark and to improve the overall performance of the portfolio when taking the specific return distribution of hedge funds into account.