Showing papers by "Paul W. Wilson published in 1998"

Sensitivity Analysis of Efficiency Scores: How to Bootstrap in Nonparametric Frontier Models

Abstract: Efficiency scores of production units are generally measured relative to an estimated pro-duction frontier. Nonparametric estimators (DEA, FDH,···) are based on a finite sample of observed production units. The bootstrap is one easy way to analyze the sensitivity of efficiency scores relative to the sampling variations of the estimated frontier. The main point in order to validate the bootstrap is to define a reasonable data-generating process in this complex framework and to propose a reasonable estimator of it. This paper provides a general methodology of bootstrapping in nonparametric frontier models. Some adapted methods are illustrated in analyzing the bootstrap sampling variations of input efficiency measures of electricity plants.

Nonparametric tests of returns to scale

Abstract: This paper discusses various statistics for testing hypotheses regarding returns to scale in the context of nonparametric models of technical eciency. In addition, the paper presents bootstrap estimation procedures which yield appropriate critical values for the test statistics. Evidence on the true sizes and power of the various proposed tests is obtained from Monte Carlo experiments. This paper is an extension of earlier work in Simar and Wilson (1998a).

Variation In Inefficiency Among Us Hospitals

Abstract: This study analyzes the impact of policy variables and other factors on hospital technical inefficiency in the US. Distance functions are used to estimate technical efficiency for each hospital relative to contemporaneous technologies. The resulting panel of efficiency estimates are then regressed on exogenous factors to gain insight into various issues impacting the debate over health-care reform in the US. Methodologically, we extend previous work by recognizing data envelopment analysis (DEA) efficiency scores as estimates, and by dealing with a censoring problem at the estimated technology frontier. We use annual data on US hospitals operated by the US Department of Veterans Affairs (VA), state and local governments, and for-profit as well as nonprofit organizations from 1985 to 1988.

Productivity growth in industrialized countries

Abstract: Several recent papers in the American Economic Review examined important questions regarding productivity growth and its sources in industrialized countries: Fare, Grosskopf, Norris, and Zhang (FGNZ),1994 and Ray and Desli (RD), 1997. We examine two sets of issues raised by these papers, and reassess what can be learned about productivity,efficiency and technology from the data used by both papers. The first set of issues are primarily economic in nature. The Malmquist measure of efficiency change was decomposed by FGNZ into measures of "pure efficiency change" and change in scale efficiency. RD offered an alternative decomposition of the Malmquist index in which the FGNZ measure of pure efficiency change appears, but in which both the scale efficiency change and technical change measures differ. But,in RD's alternative decomposition, the component which is supposed to measure changes in returns to scale confounds the different effects of movement of production units in input/output space and changes in the shape of the technology over time. We offer in this paper an alternative decomposition which avoids this problem. The second set of issues we examine concern estimation and inference. We provide a statistical model suggested by the original framework of FGNZ which allows us to estimate confidence intervals and formally test many of the claims made by both papers. The tool we used is the bootstrap methodology of Simar and Wilson (1997a,b).

A General Methodology for Bootstraping in Nonparametric Frontier Models

Abstract: The Data Envelopment Analysis method has been extensively used in the literature to provide measures of firms’ technical efficiency. These measures allow rankings of firms by their apparent performance. The underlying frontier model is nonparametric since no particular functional form is assumed for the frontier model. Since the observations result from some data-generating process, the statistical properties of the estimated efficiency measures are essential for their interpretations. In the general multi-output, multi-input framework, the bootstrap seems to offer the only means of inferring these properties (i.e. to estimate the bias and variance, and to construct confidence intervals). This paper proposes a general methodology for bootstrapping in frontier models, extending the more restrictive method proposed in Simar and Wilson (1998a) by allowing for heterogeneity in the structure of efficiency. A numerical illustration with real data is provided to illustrate the methodology.