Tracking hedge funds returns using sparse clones
- 190 Downloads
Whether hedge fund returns could be attributed to systematic risk exposures rather than managerial skills is an interesting debate among academics and practitioners. Academic literature suggests that hedge fund performance is mostly determined by alternative betas, which justifies the construction of investable hedge fund clones or replicators. Practitioners often claim that management skills are instrumental for successful performance. In this paper, we study the risk exposure of different hedge fund indices to a set of liquid asset class factors by means of style analysis. We extend the classical style analysis framework by including a penalty that allows to retain only relevant factors, dealing effectively with collinearity, and to capture the out-of-sample properties of hedge fund indices by closely mimicking their returns. In particular, we introduce a Log-penalty and discuss its statistical properties, showing then that Log-clones are able to closely track the returns of hedge fund indices with a smaller number of factors and lower turnover than the clones built from state-of-art methods.
KeywordsStyle analysis Hedge fund replication Log-penalty regression LASSO Alternative betas
We would like to thank the two anonymous referees and the Associate Editor for providing us with constructive comments that have improved the quality of our paper. Sandra Paterlini acknowledges financial support from CRoNos COST Action IC1408.
- Akaike, H. (1973). Information theory and an extension of the maximum likelihood principle. In B. N. Petrox, F. Caski F. (Eds.), Proceedings of the second international symposium on information theory (pp. 267–281). Budapest: Akademiai Kiado.Google Scholar
- BarclayHedge. (2016). Hedge fund industry-assets under management, 2016. http://www.barclayhedge.com/research/indices/ghs/mum/HF_Money_Under_Management.html.
- Fung, W., & Hsieh, D. A. (2007). Will hedge funds regress towards index-like products? Journal of Investment Management, 5(2), 46–65.Google Scholar
- Gotoh, J., & Takeda, A. (2011). On the role of norm constraints in portfolio selection. Computational Management Science, 5, 1–31.Google Scholar
- Hasanhodzic, J., & Lo, A. W. (2007). Can hedge-fund returns be replicated?: The linear case. Journal of Investment Management, 5(2), 5–45.Google Scholar
- Jaeger, L. (2008). Alternative beta strategies and hedge fund replication. Chichester: Wiley.Google Scholar
- Murphy, K . P. (2012). Machine learning: A probabilistic perspective. Cambridge: The MIT Press.Google Scholar
- Roncalli, T. (2014). Introduction to risk parity and budgeting. Financial mathematics series. London: Chapman & Hall/CRC.Google Scholar
- Sharpe, W. F. (1992). Asset allocation: Management style and performance measurement. Journal of Portfolio Management, 18(2), 7–19. Reprinted with permission from The Journal of Portfolio Management, Winter.Google Scholar
- Weber, and V., Peres, F. (2013). Hedge fund replication: Putting the pieces together. Journal of Investment Strategies, 3(1), 61–119.Google Scholar
- Weston, J., Elisseeff, A., & Schoelkopf, B. (2003). Use of the zero-norm with linear models and Kernel methods. Journal of Machine Learning Research, 3, 1439–1461.Google Scholar
- World Bank. (2015). Gross domestic product, March 2015. http://data.worldbank.org/indicator/NY.GDP.MKTP.CD.