he financial management field has witnessed significant developments in recent years to help decision makers, managers and investors, to made optimal decisions. In this regard, the institutions investment strategies and their evaluation methods continuously change with the rapid transfer of information and access to the fi- nancial data. When information is available as several inputs and output factors, the data envelopment analysis (DEA) applies to calculate the efficiency of com- panies. Distinguishing efficient companies from inefficient ones, makes it possi- ble for the financial managers to select suitable portfolios. The discriminating power of DEA depends on the number of companies under evaluation and the number of inputs and outputs. When the number of inputs and outputs are high compared to the number of units, most of the units will be evaluated as efficient, thus the discriminating power of DEA decreases and the results are not reliable. To deal with this problem, the Quick-Reduct algorithm of the rough set theory (RST) was used in this study to reduce inputs or outputs. It should be noted that the advantage of this algorithm is its ability to use negative data.
