In this paper, the financial criteria taken into account are the expected return and the difference between the returns of the portfolio and a pre-specified benchmark index i.e. a strategy of tracking error (TE) is followed. Moreover, we assume that the investor¡¯s preferences about SEE features of the portfolio are imprecisely known. In order to model these flexible preferences we propose to use fuzzy decision making. The multidimensional nature of the problem leads us to work with techniques of multiple criteria decision making (MCDM), namely goal programming (GP), and the incomplete information is handled by a fuzzy robust approach. The proposed fuzzy goal programming (FGP) model is applied to a database of UK mutual funds.