Abstract

It is shown that the finite-order universal portfolios generated by independent discrete random variables are constant rebalanced portfolios. The case where the universal portfolios are generated by the moments of the joint Dirichlet distribution is studied. The performance of the low-order Dirichlet universal portfolios on some stock-price data set is analyzed. It is demonstrated that the performance is comparable and in some cases outperform the moving-order Cover-Ordentlich universal portfolios with faster implementation time and higher wealth achieved.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call