Electronic International Standard Serial Number (EISSN)
1572-9974
abstract
The standard approach to portfolio management consists of determining the best trade-off between risk and return. This usually results in portfolios concentrated in a small collection of the available assets. To offset this one can start with a well-diversified portfolio, to be regarded as a benchmark, then search for a portfolio that improves upon its return. The main issue is that the search has to be carried out in a neighborhood in which diversification is maintained and the risk is under control. Here we examine three ways of defining the neighborhood of the well-diversified portfolio, and see that the nature of the optimal portfolio within that neighborhood changes considerably. As there is no universal choice for the distance between portfolios, this poses an interesting and hard problem for the portfolio manager.
Classification
subjects
Business
Economics
keywords
well diversifed portfolio; benchmark tracking; risk controlled optimal portfolio; maximum entropy in mean for linear programming problems