Analysis of Kellyoptimal portfolios.pdfVIP

Analysis of Kelly-optimal portfolios Paolo Laureti, Mat´uˇs Medo, Yi-Cheng Zhang Department of Physics, University of Fribourg, 8 Chemin du Mus´ee 3, 1700 Fribourg, Switzerland 0 0 2 g u Abstract A We investigate the use of Kelly’s strategy in the construction of an optimal portfolio 7 of assets. For lognormally distributed asset returns, we derive approximate analytical 2 results for the optimal investment fractions in various settings. We show that when mean returns and volatilities of the assets are small and there is no risk-free asset, the Kelly- ] h optimal portfolio lies on Markowitz Efficient Frontier. Since in the investigated case the p Kelly approach forbids short positions and borrowing, often only a small fraction of the - c available assets is included in the Kelly-optimal portfolio. This phenomenon, that we call o condensation, is explored in detail. s . s c 1 Introduction i s y h The construction of an efficient portfolio aims at maximising the investor’s capital, or its p return, while minimising the risk of unfavourable events. This problem has been pioneered by [ Markowitz in [1], where the Mean-Variance (M-V) efficient portfolio has been introduced: it 2 minimizes the portfolio variance for any fixed value of its expected return. Since this rule can v be only justified under somewhat unrealistic assumptions (namely a quadratic utility function 1 7 or a normal distribution of returns, in add