Performancemåling af nordiske hedgefonde

Christian Lønborg Thomsen & Christian Malling

Student thesis: Master thesis


The objective of this thesis is to cover the main characteristics of the Nordic hedge funds and thereby making it possible to determine suitable performance measures. Furthermore fund manager skills are evaluated, based upon regressions calculating alpha. Finally we seek to find fund specific explanatory factors to explain the presence of alpha. Performance measurement of Nordic hedge funds is not a well documented subject. Previous studies of appropriate measures have primarily been focussing on American fund returns. The Nordic hedge fund industry is very young compared to the global industry. This is primarily due to the legislature in the Nordic countries, which has not allowed the creation of hedge funds until recent years. Traditional performance measures are based on the assumption that returns are normally distributed. Subsequently risk is solely defined as volatility or market beta, meaning risk will be significantly underestimated unless the assumption of normal distribution is fulfilled. In this thesis it is shown that the returns of hedge funds are in fact not normally distributed making the results of traditional performance measures questionable. We therefore find that measuring performance should be based on non-normality. However we also find that the overall ranking of the funds is not affected by the choice of measurement unless there are substantial deviations from normality. The nature of hedge funds is complex. The use of equities, future contracts, fixed income and cash makes it hard to determine appropriate benchmarks. If improper benchmarks are used to explain manager skill results are likely to become too ambiguous. Multi-factor regressions are used to calculate alpha and to explain manager skill. We find that the use of several benchmarks as variables increase the explanatory power of the estimated alpha. In general the funds seem to outperform the market, but the low number of statistically significant alphas also indicates that the factor model is not strong enough to fully explain the fund returns. Finally, to find fund specific factors explaining the performance of hedge funds, we included measures such as the age of the fund, assets under management, the fee structure and redemption period. We find that the factors: Age, performance fee and redemption period are statistically significant. However, as all of the results have an explanatory power below 10% we find the factors unsuitable for performance evaluation. Despite low explanatory factors, which we attribute to the lack of data from the Nordic hedge funds, we overall conclude that the Nordic hedge funds have generated attractive results during the period investigated.

EducationsMSc in Finance and Accounting, (Graduate Programme) Final Thesis
Publication date2008
Number of pages122