Statistical inference in high dimensional DEA model

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

The paper aims to find a methodology to perform statistical inference for the estimation of efficiencies in the high dimensional data envelopment analysis (DEA) framework. This begins with the examination on the statistical model of the point estimator (mPCA-DEA) that eases the curse of dimensionality. To estimate the asymptotic distribution of mPCA-DEA, a double-smooth bootstrap method is adapted. To avoid unnecessary computational burden, an independence test in the distribution of efficiency is applied before bootstrapping the efficiency estimates. A numerical illustration with Malaysia car market data is used to demonstrate the methodology. It is shown that when the independence assumption holds, a double-smooth homogeneous bootstrap gives efficient inference. On the other hand, the heterogeneous bootstrap susceptibly identifies the outliers and the facets of frontier that are less dense when the independence assumption fails. These results substantiate the need to examine the asymptotic distribution of the mPCA-DEA estimator in the high dimensional framework so much so to provide meaningful interpretations. Nonetheless, the proposed methodology cannot totally overcome the curse of dimensionality in the nonparametric estimator. To further improve the estimation quality, researchers are encouraged to curtail the attributes or increase the sample size.

Original languageEnglish
Pages (from-to)17-33
Number of pages17
JournalInternational Journal of Applied Mathematics and Statistics
Volume29
Issue number5
Publication statusPublished - 2012

Fingerprint

Data envelopment analysis
Data Envelopment Analysis
High-dimensional Data
Statistical Inference
Curse of Dimensionality
Bootstrap
Asymptotic distribution
Methodology
Independence Test
Estimator
Malaysia
Nonparametric Estimator
Bootstrap Method
Bootstrapping
Facet
Model
Estimate
Statistical Model
Outlier
Sample Size

Keywords

  • Bootstrap
  • Data envelopment analysis
  • Independence

ASJC Scopus subject areas

  • Applied Mathematics

Cite this

Statistical inference in high dimensional DEA model. / Yap, G. L C; Ismail, Wan Rosmanira; Isa, Zaidi.

In: International Journal of Applied Mathematics and Statistics, Vol. 29, No. 5, 2012, p. 17-33.

Research output: Contribution to journalArticle

@article{424cfc12f1fa4bfaacdcc78d557aef31,
title = "Statistical inference in high dimensional DEA model",
abstract = "The paper aims to find a methodology to perform statistical inference for the estimation of efficiencies in the high dimensional data envelopment analysis (DEA) framework. This begins with the examination on the statistical model of the point estimator (mPCA-DEA) that eases the curse of dimensionality. To estimate the asymptotic distribution of mPCA-DEA, a double-smooth bootstrap method is adapted. To avoid unnecessary computational burden, an independence test in the distribution of efficiency is applied before bootstrapping the efficiency estimates. A numerical illustration with Malaysia car market data is used to demonstrate the methodology. It is shown that when the independence assumption holds, a double-smooth homogeneous bootstrap gives efficient inference. On the other hand, the heterogeneous bootstrap susceptibly identifies the outliers and the facets of frontier that are less dense when the independence assumption fails. These results substantiate the need to examine the asymptotic distribution of the mPCA-DEA estimator in the high dimensional framework so much so to provide meaningful interpretations. Nonetheless, the proposed methodology cannot totally overcome the curse of dimensionality in the nonparametric estimator. To further improve the estimation quality, researchers are encouraged to curtail the attributes or increase the sample size.",
keywords = "Bootstrap, Data envelopment analysis, Independence",
author = "Yap, {G. L C} and Ismail, {Wan Rosmanira} and Zaidi Isa",
year = "2012",
language = "English",
volume = "29",
pages = "17--33",
journal = "International Journal of Applied Mathematics and Statistics",
issn = "0973-1377",
publisher = "Centre for Environment Social and Economic Research",
number = "5",

}

TY - JOUR

T1 - Statistical inference in high dimensional DEA model

AU - Yap, G. L C

AU - Ismail, Wan Rosmanira

AU - Isa, Zaidi

PY - 2012

Y1 - 2012

N2 - The paper aims to find a methodology to perform statistical inference for the estimation of efficiencies in the high dimensional data envelopment analysis (DEA) framework. This begins with the examination on the statistical model of the point estimator (mPCA-DEA) that eases the curse of dimensionality. To estimate the asymptotic distribution of mPCA-DEA, a double-smooth bootstrap method is adapted. To avoid unnecessary computational burden, an independence test in the distribution of efficiency is applied before bootstrapping the efficiency estimates. A numerical illustration with Malaysia car market data is used to demonstrate the methodology. It is shown that when the independence assumption holds, a double-smooth homogeneous bootstrap gives efficient inference. On the other hand, the heterogeneous bootstrap susceptibly identifies the outliers and the facets of frontier that are less dense when the independence assumption fails. These results substantiate the need to examine the asymptotic distribution of the mPCA-DEA estimator in the high dimensional framework so much so to provide meaningful interpretations. Nonetheless, the proposed methodology cannot totally overcome the curse of dimensionality in the nonparametric estimator. To further improve the estimation quality, researchers are encouraged to curtail the attributes or increase the sample size.

AB - The paper aims to find a methodology to perform statistical inference for the estimation of efficiencies in the high dimensional data envelopment analysis (DEA) framework. This begins with the examination on the statistical model of the point estimator (mPCA-DEA) that eases the curse of dimensionality. To estimate the asymptotic distribution of mPCA-DEA, a double-smooth bootstrap method is adapted. To avoid unnecessary computational burden, an independence test in the distribution of efficiency is applied before bootstrapping the efficiency estimates. A numerical illustration with Malaysia car market data is used to demonstrate the methodology. It is shown that when the independence assumption holds, a double-smooth homogeneous bootstrap gives efficient inference. On the other hand, the heterogeneous bootstrap susceptibly identifies the outliers and the facets of frontier that are less dense when the independence assumption fails. These results substantiate the need to examine the asymptotic distribution of the mPCA-DEA estimator in the high dimensional framework so much so to provide meaningful interpretations. Nonetheless, the proposed methodology cannot totally overcome the curse of dimensionality in the nonparametric estimator. To further improve the estimation quality, researchers are encouraged to curtail the attributes or increase the sample size.

KW - Bootstrap

KW - Data envelopment analysis

KW - Independence

UR - http://www.scopus.com/inward/record.url?scp=84866760909&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84866760909&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:84866760909

VL - 29

SP - 17

EP - 33

JO - International Journal of Applied Mathematics and Statistics

JF - International Journal of Applied Mathematics and Statistics

SN - 0973-1377

IS - 5

ER -