Fuzzy DEA

Profitability as the most important factor in decision-making, has always been considered by stake­holders in the company's profitability. Also can be a basis for evaluating the performance of the managers. The ability to predict the profitability can be very useful to help decision-makers. That's why one of the most important issues is the expected profitability. The importance of these forecasts depends on the amount of misalignment with reality. The amount of deviation is less than the forecast of higher accuracy. Although there are various methods for predicting but the use of artificial intelligence techniques is increasing due to fewer restriction. The aim of this study is to evaluate the predictive power of profitability using DEA and neutral network, to enhance the decision-making users of 2012 to 2015of 7 premier financial ratios were used as independent variables. Test results show that both of ANN and DEA have ability to forecast profitability and given that neutral network prediction accuracy is higher than the DEA, the model predict better the profitability of companies.

The main goal of this paper is to propose a new approach for efficiency measurement and ranking of stocks. Data envelopment analysis (DEA) is one of the popular and applicable techniques that can be used to reach this goal. However, there are always concerns about negative data and uncertainty in financial markets. Since the classical DEA models cannot deal with negative and imprecise values, in this paper, possibilistic range directional measure (PRDM) model is proposed to measure the efficiencies of stocks in the presence of negative data and uncertainty with input/output parameters. Using the data from insurance industry, this model is also implemented for a real case study of Tehran stock exchange (TSE) in order to analyse the performance of the proposed method.