Hedge Funds : Alternative Investment Strategies and Portfolio Models

The work covers a variety of aspects, with the four main chapters combining qualitative and quantitative information on the hedge fund industry. Generally speaking, Chapters 2, 3, and the first part of Chapter 4 provide qualitative hedge fund research while quantitative aspects are discussed in the second part of Chapter 4 and in Chapter 5. The basic characteristics of hedge funds and alternative investments in general are outlined in Chapter 2. In this chapter, hedge fund industry standards and the characteristics typically exhibited by hedge funds are described in detail and compared to those in the traditional investment universe. The chapter also concentrates on the definition of hedge funds and the alternative investment universe. After discussing the special features of the hedge fund industry, we analyze which sources of risk may justify the historically very attractive returns achieved by the different hedge fund strategies. Although the relevant literature contains few systematic overviews of individual hedge fund strategies and the corresponding trades, such an analysis of individual strategies and funds of hedge funds is essential when investment in hedge funds is considered. In Chapter 3 systematic and unsystematic risk factors in hedge fund investing and the risk premia captured by hedge fund strategies are introduced. Furthermore, the relevant trades and the corresponding risk factors for several classical hedge fund strategies are described. Since new investors in hedge funds typically choose the fund of hedge funds structure instead of direct investment in single hedge funds, Chapter 3 also analyzes the fund of hedge funds structure. This way of structuring hedge fund investments offers investors some advantages but also has some drawbacks. Chapter 4 addresses the need for reliable hedge fund return data. When investors consider allocating money to hedge funds they form an expectation of the development of future returns. Since historical data is usually their starting point in this, Chapter 4 examines in detail the most important hedge fund index data providers and the relevant indices and takes a closer look at the different hedge fund databases and the different index methodologies. After discussing several data biases and their implications, the statistical properties of fund of hedge funds indices are analyzed. Such an analysis of historical behavior is a vital issue for an investor interested in the quantitative modeling of the risk return characteristic and the impact of hedge funds. This in-depth analysis also forms the starting point for the analysis of a portfolio including hedge funds. The implications of the statistical properties of hedge fund index returns for portfolio selection are discussed in the first sections of Chapter 5. We introduce an alternative one-period modeling approach for portfolio returns that is flexible enough to capture these return characteristics. While Markowitz's classical portfolio selection framework involves rather strict assumptions about the structure of asset returns or utility functions, the approach presented in this work is very flexible. In a first step we model the marginal return distributions of the individual assets in order to reproduce the univariate return characteristics of the assets under consideration. In a second step, we focus on the dependence structure of the asset returns in the portfolio. This modeling of dependencies with multivariate copula functions delivers more realistic results than the classical portfolio selection framework. Combining the marginal return distributions and the dependence structures delivers a multivariate model for asset returns. This model is applied in a case study which analyzes a traditional portfolio in which hedge funds are included. This is a major issue for hedge fund investors and, with the model for portfolio returns at hand, we are able to derive optimal allocations to asset classes coupled with an optimization with respect to different risk measures. Chapter 6 concludes with a brief summary of the most important results. In addition, we make suggestions for future research.Gegenstand der Arbeit sind verschiedene qualitative und quantitative Aspekte von Hedgefonds, welche für ein umfassendes Verständnis der Anlageklasse von zentraler Bedeutung sind. Die Arbeit enthält daher zunächst eine Einführung zu Hedgefonds im Allgemeinen sowie in die Charakteristika dieser Anlagealternative. Das dritte Kapitel beschäftigt sich zudem mit einzelnen Hedgefonds-Handelsstrategien, den zugehörigen Risikofaktoren und den spezifischen Eigenschaften von Dachhedgefonds. Nach der Untersuchung der Renditequellen wird im vierten Kapitel das verfügbare Hedgefonds-Indexuniversum näher analysiert. Hierbei stehen sowohl die statistischen Eigenschaften als auch die in der wissenschaftlichen Literatur diskutierten Datenverzerrungen im Zentrum der Analyse. Dies stellt die Basis für das anschließend eingeführte Portfoliomodell dar. Das vorgestellte Portfoliomodell ist in der Lage nicht-normalverteilten Renditen einzelner Anlageklassen zu berücksichtigen sowie komplexe nicht-lineare Abhängigkeitsstrukturen zwischen einzelnen Anlageklassen über Copula-Funktionen abzubilden. Die durch das Portfoliomodell gegebene Möglichkeit der Abbildung der ermittelten Hedgefonds-Charakteristika wird in einer abschließenden Fallstudie unterstrichen