Joint Variable Selection for Data Envelopment Analysis via Group Sparsity

28 Pages Posted: 10 Mar 2014

See all articles by Zhiwei Qin

Zhiwei Qin

DiDi Research America; affiliation not provided to SSRN

Irene Song

Columbia University - Department of Industrial Engineering and Operations Research (IEOR)

Date Written: March 9, 2014

Abstract

This study develops a data-driven group variable selection method for data envelopment analysis (DEA), a non-parametric linear programming approach to the estimation of production frontiers. The proposed method extends the group Lasso (least absolute shrinkage and selection operator) designed for variable selection on (often predefined) groups of variables in linear regression models to DEA models. In particular, a special constrained version of the group Lasso with the loss function suited for variable selection in DEA models is derived and solved by a new tailored algorithm based on the alternating direction method of multipliers (ADMM). This study further conducts a thorough evaluation of the proposed method against two widely used variable selection methods -- the efficiency contribution measure (ECM) method and the regression-based (RB) test -- in DEA via Monte Carlo simulations. The simulation results show that our method provides more favorable performance compared with its benchmarks.

Keywords: Data envelopment analysis (DEA), Group Lasso, Lasso, Group variable selection, Alternating direction method of multipliers (ADMM)

Suggested Citation

Qin, Zhiwei and Song, Irene, Joint Variable Selection for Data Envelopment Analysis via Group Sparsity (March 9, 2014). Available at SSRN: https://ssrn.com/abstract=2406690 or http://dx.doi.org/10.2139/ssrn.2406690

Zhiwei Qin (Contact Author)

DiDi Research America ( email )

450 National Ave
Mountain View, CA 94043
United States

affiliation not provided to SSRN

Irene Song

Columbia University - Department of Industrial Engineering and Operations Research (IEOR) ( email )

331 S.W. Mudd Building
500 West 120th Street
New York, NY 10027
United States

Register to save articles to
your library

Register

Paper statistics

Downloads
57
Abstract Views
404
rank
365,668
PlumX Metrics