Partial index tracking: satisfying different investment profiles with the same subset of stocks

Abstract

Pasive portfolio management aims to minimize the unsystematic risk of the portfolios by imitating the behaviour of a stock index. Partial index tracking enables passive portfolio management by only considering a subset of the stocks included in the index obtaining a substantial cost reduction compared with full index tracking. In the literature, three criteria are usually employed to undertake partial index tracking: tracking error variance, portfolio variance and expected return. We propose an additional criterion to be considered, the frontier curvature. The main advantage when using this new criterion for portfolio selection is that a manager can satisfy different investment profiles with the same subset of stocks.