Single manager hedge funds - aspects of classification and diversification

[摘要] A persistent problem for hedge fund researchers presents itself in the form ofinconsistent and diverse style classifications within and across database providers. Forthis paper, single-manager hedge funds from the Hedge Fund Research (HFR) andHedgefund.Net (HFN) databases were classified on the basis of a common factor,extracted using the factor axis methodology. It was assumed that the returns of allsample hedge funds are attributable to a common factor that is shared across hedgefunds within one classification, and a specific factor that is unique to a particular hedgefund. In contrast to earlier research and the application of principal component analysis,factor axis has sought to determine how much of the covariance in the dataset is due tocommon factors (communality). Factor axis largely ignores the diagonal elements of thecovariance matrix and orthogonal factor rotation maximises the covariance betweenhedge fund return series.In an iterative framework, common factors were extracted until all return series weredescribed by one common and one specific factor. Prior to factor extraction, the serieswas tested for autoregressive moving-average processes and the residuals of suchmodels were used in further analysis to improve upon squared correlations as initialfactor estimates. The methodology was applied to 120 ten-year rolling estimationwindows in the July 1990 to June 2010 timeframe. The results indicate that the numberof distinct style classifications is reduced in comparison to the arbitrary self-selectedclassifications of the databases. Single manager hedge funds were grouped in portfolioson the basis of the common factor they share. In contrast to other classificationmethodologies, these common factor portfolios (CFPs) assume that some unspecifiedindividual component of the hedge fund constituents' returns is diversified away and thatsingle manager hedge funds should be classified according to their common returncomponents. From the CFPs of single manager hedge funds, pure style indices werecreated to be entered in a multivariate autoregressive framework.For each style index, a Vector Error Correction model (VECM) was estimated todetermine the short-term as well as co-integrating relationship of the hedge fund series with the index level series of a stock, bond and commodity proxy. It was postulated thata) in a well-diversified portfolio, the current level of the hedge fund index is independentof the lagged observations from the other asset indices; and b) if the assumptions of theEfficient Market Hypothesis (EMH) hold, it is expected that the predictive power of themodel will be low. The analysis was conducted for the July 2000 - June 2010 period.Impulse response tests and variance decomposition revealed that changes in hedgefund index levels are partially induced by changes in the stock, bond and currencymarkets. Investors are therefore cautioned not to overemphasise the diversificationbenefits of hedge fund investments. Commodity trading advisors (CTAs) / managedfutures, on the other hand, deliver diversification benefits when integrated with anexisting portfolio.The results indicated that single manager hedge funds can be reliably classified usingthe principal factor axis methodology. Continuously re-balanced pure style indexrepresentations of these classifications could be used in further analysis. Extensivemultivariate analysis revealed that CTAs and macro hedge funds offer superiordiversification benefits in the context of existing portfolios. The empirical results are ofinterest not only to academic researchers, but also practitioners seeking to replicate themethodologies presented.

[发布机构] Stellenbosch University