However, stochastic frontier analysis (SFA) is predominantly
used for estimating technical efficiency studies in the aquaculture
industry (Appendix A), perhaps because DEA attributes all deviations
from the production frontier to technical inefficiency,thereby
making it an inappropriate technique in some sectors, especially
in agriculture, in which the data collection process is sensitive to
stochastic noise and other measurement errors (Coelli et al., 2005).
This shortcoming of DEA led Simar and Wilson (1998, 2000) to
propose a technique which allows the construction of confidence
intervals for DEA technical efficiency scores with the help of bootstrapping
procedures. The reason for bootstrapping is to estimate
the bias-corrected technical efficiency (BCTE), which is more accurate
estimates of efficiency scores than the conventional DEA.