Abstract
Data envelopment analysis (DEA) is a popular non-parametric approach to examine performance and productivity of airlines; however, it could not provide statistical information such as confidence intervals on the estimated efficiency scores. We combined stochastic frontier analysis and DEA into a single framework to disentangle noise and ‘pure’ inefficiency from the DEA inefficiency scores and accordingly provide confidence intervals for the estimated efficiency scores. Monte-Carlo simulation verified that our novel model is a good alternative for the conventional DEA as well as the bootstrap DEA. Empirical application using Asia-Pacific airlines’ data (2008‒2015) shows that after accounting for the ‘pure’ random errors, the sampled Asia-Pacific airlines performed well during the study period but their ‘pure’ efficiency was declining, hence, there is still room for improvement.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have