Estimating Vessel Efficiency Using a Bootstrapped Data Envelopment Analysis Model

Abstract

Technical efficiency, which measures how well a firm transforms inputs into outputs, gives fishery managers important information concerning the economic status of the fishing fleet, and how regulations may be impacting vessel profitability. Data envelopment analysis (DEA), and the stochastic production frontier (SPF) have emerged as preferred methods to estimate efficiency in fisheries. Although each of the approaches has strengths and weaknesses, DEA has often been criticized because it is "deterministic" and fails to account for noise in the data. This paper presents a method for examining the underlying statistical structure of DEA models using bootstrap methods, and readily available software. The approach is then applied to a case study of the U.S. mid-Atlantic sea scallop dredge fleet. Results show that the 95% confidence interval for technically efficient output is well above the maximum sustained yield (MSY) level of output.Short Title: Measuring Vessel Efficiency

Economic Benefits of Management Reform in the Gulf of Mexico Grouper Fishery: A Semi-parametric Analysis

“…Studies examining the performance of existing programs are increasingly available, e.g., Grafton, Squires and Fox (2000). 5 Single-step nonparametric frontier models have been used extensively to study economic performance and capacity utilization in commercial fisheries (e.g., Kirkley et al, 2002;Weninger and Waters, 2003;Walden, 2006). Färe et al (1994), and Coelli et al (1998) provide exhaustive reviews of nonparametric frontier models.…”

Traffic fatality indicators in Brazil: State diagnosis based on data envelopment analysis research

“…In this context, the general criticism relies on the deterministic nature of DEA estimators, centered on the argument that the technique does not account for uncertainties in the data, being merely a point estimator from which no statistical inference can be derived. When running a DEA model, a production frontier based on the provided input information is constructed and is used as a reference to compute the set of scores; thus, it is not difficult to conclude that the configuration of such a frontier is susceptible to variations of the input information (Löthgren, 1998;Simar and Wilson, 1998;Walden, 2006).…”

Bootstrapping DEA Scores for Road Safety Strategic Analysis in Brazil

“…These scores are resampled with replacement according to the empirical distribution, generating the set 1s* , ..., 1n* and, afterwards, they should be smoothed, forming the set 1s* 1n* . Smoothening is the term attributed to the process of incorporating the properties of the original set of index scores into the resampled set, through the use of reflection methods (12,13), represented by the application of Eq. (4) to (6) set y 2i,s , ..., y 2p,n , which is normalized prior to computing the second set of index scores.…”