Factors That Influence the Accuracy of Estimates of Capacity and
Capacity Utilization Derived from Data Envelopment Analysis
By Daniel Holland and S. Todd Lee
ABSTRACT
Data Envelopment Analysis (DEA) has been proposed as a tool to measure technical capacity for fishing vessels and fleets. However, there are several factors that can lead to bias in DEA estimates of fishing capacity. Allowing for variable returns to scale will tend to reduce DEA estimates of capacity. Random variation in commercial fishery catch data can lead to upward bias. This research uses Monte-Carlo simulations to investigate possible biases in DEA estimates of technically efficient output and capacity output under alternative model specifications and data generating processes. We demonstrate a simple method of reducing noise induced bias when panel data is available. We show that DEA estimates of fishing capacity are highly sensitive to both model specification and noise, but that estimates of capacity utilization are more robust.
KEYWORDS: Capacity, Capacity Utilization, Data Envelopment Analysis, Monte Carlo Analysis
 View Capacity: Data Envelope Analysis Session
|
