Variable selection in Data Envelopment Analysis

The selection of inputs and outputs in Data Envelopment Analysis (DEA) is regarded as an important step that is normally conducted before the DEA model is implemented. In this paper, we introduce cardinality constraints directly into the DEA program in order to select the relevant inputs and outputs automatically, without any previous statistical analysis, heuristic decision making or expert judgement (though our method is not incompatible with these other approaches and indeed may help to choose among them). The selection of variables is obtained solving a mixed integer linear program (MILP) which specifies the maximal number of variables to be used. The computational time of the program is fast in all practical situations. We explore the performance of the method via Monte Carlo simulations. Some empirical applications are considered in order to illustrate the usefulness of the method.

[1]  Angelo Zago,et al.  Large Courts, Small Justice! On the inefficiency and optimal structure of the Italian Justice sector , 2016 .

[2]  Dolores Romero Morales,et al.  Feature Selection in Data Envelopment Analysis: A Mathematical Optimization approach , 2020, Omega.

[3]  William W. Cooper,et al.  Evaluating Program and Managerial Efficiency: An Application of Data Envelopment Analysis to Program Follow Through , 1981 .

[4]  P. W. Wilson,et al.  TESTING RESTRICTIONS IN NONPARAMETRIC EFFICIENCY MODELS , 2001 .

[5]  Larry Jenkins,et al.  A multivariate statistical approach to reducing the number of variables in data envelopment analysis , 2003, Eur. J. Oper. Res..

[6]  B. Park,et al.  A NOTE ON THE CONVERGENCE OF NONPARAMETRIC DEA ESTIMATORS FOR PRODUCTION EFFICIENCY SCORES , 1998, Econometric Theory.

[7]  A. Charnes,et al.  Some Models for Estimating Technical and Scale Inefficiencies in Data Envelopment Analysis , 1984 .

[8]  Nicole Adler,et al.  Improving discrimination in data envelopment analysis: PCA-DEA or variable reduction , 2010, Eur. J. Oper. Res..

[9]  Jesús T. Pastor,et al.  A Monte Carlo Evaluation of several Tests for the Selection of Variables in Dea Models , 2005, Int. J. Inf. Technol. Decis. Mak..

[10]  Andrew L. Johnson,et al.  Guidelines for using variable selection techniques in data envelopment analysis , 2011, Eur. J. Oper. Res..

[11]  Léopold Simar,et al.  Advanced Robust and Nonparametric Methods in Efficiency Analysis: Methodology and Applications , 2007 .

[12]  Abraham Charnes,et al.  Measuring the efficiency of decision making units , 1978 .

[13]  Gonzalo Guillén-Gosálbez,et al.  Mixed-Integer Programming Approach for Dimensionality Reduction in Data Envelopment Analysis: Application to the Sustainability Assessment of Technologies and Solvents , 2018 .

[14]  Chia-Yen Lee,et al.  LASSO variable selection in data envelopment analysis with small datasets , 2020 .

[15]  Rajiv D. Banker,et al.  Hypothesis tests using data envelopment analysis , 1996 .

[16]  John Ruggiero,et al.  Impact Assessment of Input Omission on Dea , 2005, Int. J. Inf. Technol. Decis. Mak..

[17]  T. Ueda,et al.  APPLICATION OF PRINCIPAL COMPONENT ANALYSIS FOR PARSIMONIOUS SUMMARIZATION OF DEA INPUTS AND/OR OUTPUTS , 1997 .

[18]  Janet M. Wagner,et al.  Stepwise selection of variables in data envelopment analysis: Procedures and managerial perspectives , 2007, Eur. J. Oper. Res..

[19]  Rajiv D. Banker,et al.  Efficiency Analysis for Exogenously Fixed Inputs and Outputs , 1986, Oper. Res..

[20]  Hiroshi Morita,et al.  SELECTING INPUTS AND OUTPUTS IN DATA ENVELOPMENT ANALYSIS BY DESIGNING STATISTICAL EXPERIMENTS( Operations Research for Performance Evaluation) , 2009 .

[21]  Boaz Golany,et al.  Evaluation of deregulated airline networks using data envelopment analysis combined with principal component analysis with an application to Western Europe , 2001, Eur. J. Oper. Res..

[22]  Lawrence M. Seiford,et al.  Data envelopment analysis (DEA) - Thirty years on , 2009, Eur. J. Oper. Res..

[23]  Laurens Cherchye,et al.  An Introduction to ‘Benefit of the Doubt’ Composite Indicators , 2007 .

[24]  John S. Liu,et al.  Data envelopment analysis 1978-2010: A citation-based literature survey , 2013 .

[25]  Seok-Oh Jeong,et al.  ASYMPTOTIC DISTRIBUTION OF CONICAL-HULL ESTIMATORS OF DIRECTIONAL EDGES , 2010 .

[26]  Zilla Sinuany-Stern,et al.  Combining ranking scales and selecting variables in the DEA context: the case of industrial branches , 1998, Comput. Oper. Res..

[27]  Ali Emrouznejad,et al.  Evaluation of research in efficiency and productivity: A survey and analysis of the first 30 years , 2008 .

[28]  Paul W. Wilson,et al.  Dimension reduction in nonparametric models of production , 2017, Eur. J. Oper. Res..

[29]  Emmanuel Thanassoulis,et al.  Weights restrictions and value judgements in Data Envelopment Analysis: Evolution, development and future directions , 1997, Ann. Oper. Res..

[30]  Jesús T. Pastor,et al.  A Statistical Test for Nested Radial Dea Models , 2002, Oper. Res..

[31]  Yaakov Roll,et al.  An application procedure for DEA , 1989 .

[32]  R. Färe,et al.  A note on decomposing the Malmquist productivity index by means of subvector homotheticity , 2001 .