Efficiency evaluation of fuzzy portfolio in different risk measures via DEA

In this paper, we discuss the fuzzy portfolio efficiency evaluation problem in different risk measures. Real frontier approach (RFA) is often used in portfolio performance assessment. However, the computation complexity and the real trading solution make it hard to achieve in practice. In this work, we first present three kinds of DEA (Data envelopment analysis) based fuzzy portfolio estimation models in different risk measures, i.e., possibilistic variance, possibilistic semi-variance, and possibilistic semi-absolute deviation, to evaluate the portfolio efficiency (PE). Furthermore, we carry out large amount of simulations with different sample sizes to compare our proposed models with RFA. All results demonstrate that with adequate sample size, the envelop frontier generated by our models can approximate the real effective portfolio frontier, and PE obtained by these two methods are highly related.