The present study is an attempt toward evaluating the performance of portfolios and assets selecting using modified mean-variance models by utilizing a non-parametric efficiency analysis tool, namely Data Envelopment Analysis (DEA). Huge amounts of money are being invested in financial market. As a result, portfolio performance evaluation has created a great deal of interest among people. We know that, for calculating portfolio variance measure based on mean-variance model, the covariance between each pair of the assets is not equal to zero. Consequently sharp’s single factor model is used with linear regression for efficiency evaluation in modified mean-variance models. Since the covariance between two stocks are not merely bound to the characteristics of the two stocks but these stocks are connected together through their relations to the market return, the total number of parameters that needs to be estimated is reduced.