Previous studies have indicated that investors and fund managers in financial service industry were rarely able to earn superior returns. Therefore, selecting stocks with high expected returns to beat the market seems difficult to achieve. The objective of this study is to apply data envelopment analysis (DEA) models to evaluate the efficiency of the firms and construct portfolios by selecting stocks with high efficiency. The return rates of the portfolios constructed by DEA models and market indexes were compared via empirical data analysis. Moreover, this study evaluated the effect of input and output factors to stock selection performance using DEA methods by comparing two groups of input and output factors. The results showed that the return performances of CCR and BCC models in DEA methods significantly better than the Industry Average in the empirical time period. In addition, the portfolios constructed by the second group of factors with previous literature support had better performance than the portfolios constructed by the first group of factors.