M. Markov, I. Muchnik (USA), V. Mottl, and O. Krasotkina (Russia)
Style analysis of investment portfolios, hedge funds, time varying regression, leave-one-out principle, Kalman filter.
In this paper, we review one of the most effective finan cial multi-factor models, called Returns Based Style Analy sis (RBSA), from the standpoint of its performance in de tecting dynamic factor exposures of hedge funds using only fund performance data. We analyze the shortcomings of earlier models in capturing changes in a dynamic portfolio structure and lay the groundwork for a new approach, which we call Dynamic Style Analysis (DSA). The problem is treated as that of estimating a time-varying regression model of the observed time series with the inevitable neces sity to choose the appropriate level of model volatility, ranging from the full stationarity of instant models to their absolute independence of each other. We further propose an efficient method of model estimation and introduce a novel measure of the validity Predicted 2 R that is used to select the model parameters. Using both model and real hedge fund returns we illustrate the advantages of the proposed technique in analysis of hedge funds.
Important Links:
Go Back
