Improving the accuracy of DEA efficiency analysis: a bootstrap application to the health care foodservice industry

Kriščiukaitienė

2013

Agricultural Economics

This article presents insights into the efficiency of Lithuanian family farms. The research covers the years 2004-2009 and is based on farm-level Farm Accountancy Data Network data. Bootstrapped data envelopment analysis was employed to obtain efficiency scores, whereas stochastic kernels and multivariate methods, viz. fuzzy clustering and nonparametric regression, were utilized to assess the impact of selected determinants on efficiency. Clustering analysis indicated that the efficiency change paths specific for the analyzed sample differed in both their average levels and ranges. Furthermore, it was concluded that Lithuanian agricultural policy should focus on increasing the efficiency of crop farms. Cluster analysis suggested that production subsidies might be having a negative effect on efficiency.JEL classifications: C440, C610, Q100, Q120

Section: Preliminaries For Benchmarkingmentioning

confidence: 99%

“…Bootstrapped DEA was introduced by Simar and Wilson (, , ). Assaf and Matawie () employed bootstrapped DEA to assess the efficiency of health care foodservice facilities. Aldea and Ciobanu () and Aldea et al.…”

Section: Introductionmentioning

confidence: 99%

A nonparametric analysis of the determinants of family farm efficiency dynamics in Lithuania

Kriščiukaitienė

2013

Agricultural Economics

“…Thus, regression analysis cannot be robust with this multicollinearity problem. For this reason, a double bootstrapping technique can better reveal the impact of exogenous and operational variables on efficiency scores [10,64]. However, Banker and Natarajan [65] demonstrated that two-stage DEA with ordinary least squares or Tobit outperforms one-stage parametric methods, such as translog and Cobb-Douglas, in defining the production frontier.…”

Section: Discussionmentioning

confidence: 99%

The Effects of Operational and Environmental Variables on Efficiency of Danish Water and Wastewater Utilities

Guerrini

Romano

Leardini

et al. 2015

Water

Efficiency improvement is one of three patterns a public utility should follow in order to get funds for investments realization. The other two are recourse to bank loans or to private equity and tariff increase. Efficiency can be improved, for example, by growth and vertical integration and may be conditioned by environmental variables, such as customer and output density. Prior studies into the effects of these variables on the efficiency of water utilities do not agree on certain points (e.g., scale and economies of scope) and rarely consider others (e.g., density economies). This article aims to contribute to the literature by analysing the efficiency of water utilities in Denmark, observing the effects of operational and environmental variables. The method is based on two-stage Data Envelopment Analysis (DEA) applied to 101 water utilities. We found that the efficiency of the water sector was not affected by the observed variables, whereas that of wastewater was improved by smaller firm size, vertical integration strategy, and higher population density.

2014 IEEE International Conference on Industrial Engineering and Engineering Management

“…Hence, the efficiency results in biascorrected is definitely influenced by the error portion discarded. Assaf and Matawie [28] noted that due to the upward-bias in the original estimates and the bootstrap correction in the confidence interval (95%), the original estimates lie to outside for every observation, but can close to the lower bound for the confidence interval. However, the bias corrected estimates lie inside for each observation within the confidence interval.…”

Section: B Efficiency Estimated Bias-corrected Analysis Based On Bdementioning

confidence: 96%

A Bootstrap Data Envelopment Analysis (BDEA) approach in Islamic banking sector: A method to strengthen efficiency measurement

Zakaria

Salleh

Hassan

2014

This study applied the Bootstrapping Data Envelopment Analysis (BDEA) approach to investigate the efficiency level of twelve Islamic banks in Malaysia. BDEA approach has been employed to resolve the uncertainty issue of traditional DEA measurement practice as well as small sample problem. The empirical analysis showed that in measuring Islamic banks' efficiency, there is a bias issue that BDEA approach has been able to trace, thereby providing more accurate and reliable efficiency results. We believe that this approach, not only is important for a more precise efficiency measurement but also compatible with Islamic banking concepts that differ substantially in objectives and operations from those of conventional banks. In general, this study has contributed to the efficiency study literature where the issue of accuracy and bias in examining efficiency level is crucial.