Data Envelopment Analysis-Discriminant Analysis by imprecise data for more than two groups: apply to the pharmaceutical stock companies

One of the interesting subjects that amuse the mind of researchers is surmising the correct classification of a new sample by using available data. Data Envelopment Analysis (DEA) and Discriminant Analysis (DA) can classify data by each one alone. DEA classifies as efficient and inefficient groups and DA classify by historical data. Merge these two methods is a powerful tool for classifying the data. Since, in the real world, in many cases we do not have the exact data, so we use imprecise data (e.g. fuzzy and interval data) in these cases. So, in this paper, we represent our new DEA-DA method by using Mixed-Integer Nonlinear Programming (MINLP) to classify with imprecise data to more than two groups. Then we represent an empirical example of our purpose method on the Iranian pharmaceutical stock companies' data. In our research, we divided pharmaceutical stock companies into four groups with imprecise data (fuzzy and interval data). Since, most of the classical DA models used for two groups, the advantage of the proposed model is beheld. The result shows that the model can predict and classify more than two groups (as many as we want) with imprecise data so correct.