AccScience Publishing / IJOCTA / Volume 7 / Issue 1 / DOI: 10.11121/ijocta.01.2017.00334
RESEARCH ARTICLE

Copula approach to select input/output variables for DEA

Olcay Alpay1* Elvan Akturk Hayat2
Show Less
1 Department of Statistics, Sinop University, Turkey
2 Department of Econometrics, Adnan Menderes University, Turkey
Submitted: 5 April 2016 | Accepted: 15 August 2016 | Published: 25 October 2016
© 2016 by the Author(s). This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution -Noncommercial 4.0 International License (CC-by the license) ( https://creativecommons.org/licenses/by-nc/4.0/ )
Abstract

Determination of the input/output variables is an important issue in Data  Envelopment Analysis (DEA). Researchers often refer to expert opinions in  defining these variables. The purpose of this paper is to propose a new approach  to determine the input/output variables, it is important to keep in mind that  especially when there is no any priori information about variable selection. This  new proposed technique is based on a theoretical method which is called  “Copula”. Copula functions are used for modeling the dependency structure of  the variables with each other. Also we use the local dependence function which  analyzes the point dependency of variables of copulas to define the input/output  variables. To illustrate the usefulness of the proposed approach, we conduct two  applications using simulated and real data and compare the efficiencies in DEA.  Our results show that new approach gives values close to perfection.

Keywords
Data envelopment analysis
Variable selection
Copulas
Local dependence function
Conflict of interest
The authors declare they have no competing interests.
References

[1] Charnes A., Cooper W.W. and Rhodes E.,  Measuring the Efficiency of Decision Making  Units. European Journal of Operational Research,  2(6), 429-444 (1978).

[2] Yoluk, M., Hastane Performasının Veri Zarflama  Analizi (VZA) Yöntemi İle Değerlendirilmesi.  Thesis (MSc). Atılım University (2010).

[3] Morita, H. and Avkiran, N.K., Selecting Inputs and  Outputs in Data Envelopment Analysis by  Designing Statistical Experiments. Journal of  Operational Research Society of Japan, 52, 163-173  (2009).

[4] Nelsen, R.B., An Introduction to Copulas. Springer,  New York (1999).

[5] Bairamov, I. and Kotz, S., On Local Dependence  Function for Multivariate Distributions. New  Trends in Prob. and Stat., 5, 27-44 (2000).

[6] Bairamov, I., Kotz, S. and Kozubowski, T.J., A  New Measure of Linear Local Dependence.  Statistics, 37(3), 243-258 (2003).

[7] Ramanathan, R., An Introduction to Data  Envelopment Analysis. Sage Publications, New  Delhi (2003).

[8] Luo, Y., Bi, G. and Liang, L., Input/Output  Indicator Selection for DEA Efficiency Evaluation:  An Empirical Study of Chinese Commercial  Banks. Expert System with Applications, 39, 1118- 1123 (2012). 

[9] Pastor, J.T., Ruiz J.L. and Sirvent, I., A Statistical  Test for Nested Radial Dea Models. INFORMS,  50(4), 728-735 (2002)

[10] Jenkins, L. and Anderson, M., A multivariate  statistical approach to reducing the number of  variables in data envelopment analysis. European  Journal of Operational Research, 147, 51–61  (2003).

[11] Ruggiero, J., Impact Assessment of Input Omission  on DEA. International Journal of Information  Technology and Decision Making, 4(3), 359-368,  (2005).

[12] Edirisinghe, N.C.P. and Zhang, X., Generalized  DEA Model of Fundamental Analysis and Its  Application to Portfolio Optimization. Journal of  Banking and Finance, 31, 311-335, (2007)

[13] Madhanagopal, R. and Chandrasekaran, R.,  Selecting Appropriate Variables for DEA Using  Genetic Algorithm (GA) Search Procedure.  International Journal of Data Envelopment Analysis  and Operations Research, 1(2), 28-33 (2014).

[14] Banker, R.D., Charnes, A. and Cooper W.W., Some  Models for Estimating Technical and Scale  Inefficiencies in Data Envelopment Analysis.  Management Science, 30(9), 1078-1092 (1984).

[15] Fare, R. and Grosskopf, S., and Lovell, C.A.K.,  Production Frontiers. Cambridge University Press  (1993).

[16] Martic, M.M., Novakovic, M.S. and Baggia, A.,  Data Envelopment Analysis - Basic Models and  their Utilization, Organizacija. Journal of  Management, Informatics and Human Resources,  42(2), 37-43 (2009).

[17] Cooper, W.W., Seiford, L.M. and Zhu, J.,  Handbook on Data Envelopment Analysis. Springer  New York (2011).

[18] Peňa, V.H. and Ibragimov, R., Sharakhmetov, S.,  Characterizations of Joint Distributions, Copula  Information, Dependence and Decoupling, with  Applications to Time Series. IMS Lecture NotesMonograph Series, 2nd Lehmann SymposiumOptimality, 49, 183-209 (2006).

[19] Bairamov, I. and Kotz, S., Dependence Structure  and Symmetry of Huang-Kotz FGM Distributions  and Their Extensions. Metrica, 56, 55-72 (2002).

[20] Bairamov, I., Kotz, S. and Bekci, M., New  Generalized Farlie-Gumbel-Morgenstern  Distributions and Concomitants of Order Statistics.  Journal of Applied Statistics, 28(5), 521-536  (2001).

[21] Najjari, V., Yeni Arşimedyan Kapula Aileleri ve  Finans Alanında Bir Uygulama. Thesis (PhD). Gazi  University (2014).

[22] Vandenhende, F., and Lambert, P., Local  dependence estimation using semiparametric  Archimedean copulas. The Canadian Journal of  Statistics, 33(3), 377-388, (2005).

Share
Back to top
An International Journal of Optimization and Control: Theories & Applications, Electronic ISSN: 2146-5703 Print ISSN: 2146-0957, Published by AccScience Publishing