Estimation of cardinality constrained portfolio efficiency via segmented DEA in Iran Capital Market

The purpose of this study is to determine the efficiency limit and evaluate the performance of the stock portfolio in issues with numerical limitations in the Tehran Stock Exchange and IRAN Farabors in year 1398, using data envelopment analysis method. The efficient frontier in the issue of stock portfolio with numerical constraints is discrete and asymmetric. It is not possible to use the data envelopment analysis model. In this study, the efficient boundary is categorized into several continuous boundaries and the "data envelopment analysis" model is used. For this purpose, first the beginning and end points of each continuous boundary are determined as "data segment points" based on the proposed algorithm, and then the data envelopment analysis model is used for each continuous interval. 188 stock exchange and OTC symbols listed in Tehran Stock Exchange and IRAN Farabors were examined in 1398, and with the limitation of selecting 5 to 10 stocks from 188 stocks based on different weights (total stock weight equal to 1), the efficient limit of the stock portfolio was determined. Due to the lack of fast and accurate computational methods in determining the efficiency limit of stock portfolio with numerical constraints, data envelopment analysis method to determine the efficiency limit and evaluate stock portfolio performance of particle swarm algorithm, genetic algorithm and imperialist competition algorithm And the efficient border can be drawn and used.