A hybrid Model on the Basis of Data Envelopment Analysis and Data Mining Techniques to Analyze the Investment Behavior in Stock Exchange: A Real Case Study in Tehran Stock Exchange
Subject Areas : International Journal of Finance, Accounting and Economics StudiesSaiedeh Molla Hosseinagha 1 , Kaveh Khalili-Damghani 2
1 - Department of Industrial Engineering,
South-Tehran Branch,
Islamic Azad University, Tehran, Iran
2 - Department of Industrial Engineering,
South-Tehran Branch, Islamic Azad University, Tehran, Iran
Keywords: Clustering, Efficiency, Data envelopment analysis (DEA), data mining, Keywords: Stock exchange,
Abstract :
Stock exchange, which is the core of investment market, transfers scattered resources to the productive sector of the service and industry. The analysis of stock exchange has attracted lots of researches in recent decades. The issue concern both of securities and investors. More formally both sides of investment (i.e., behavior of investors and fluctuations of returns on securities) form the markets. In this paper, a hybrid model on the basis of data envelopment analysis and data mining techniques is proposed to analyze the investment behavior in stock exchange. The proposed approach is applied in Tehran stock exchange. First, the financial data of investments on selective companies during 2013-2014 are analyzed using data mining approaches to recognize the behavioral patterns of investors and securities.Second, customers are clustered in 3 selling and 4 buying groups using data mining techniques. Then, the efficiency of active companies in stock exchange is evaluated using and input-oriented Data Envelopment Analysis model considering variable returns-to-scale assumption. Data mining techniques and DEA model are codified using MATLAB and LINGO software, respectively. The obtained results indicate that in 2013 and 2014,nine and six companies are efficient, respectively.Among these companies, six companies are efficient in both years. The most visited companies in reference sets are determined during periods of this study. The reference set of inefficient companies are determined in order to project them toward efficient frontier. The results of this study can provide insightful vision to investors in order to illustrate the behavior of previous investors and securities.