A structural vector autoregressive model of technical efficiency and delays with an application to Chinese airlines.

This study reports on the performance assessment of Chinese airlines from 2006 to 2014 using a stochastic distance function where technical efficiency and a measure of flight delays follow a joint structural autoregressive process. This model is used to investigate whether technical efficiency causes flight punctuality or the other way around. The model, however, yields a non-trivial likelihood function and is not amenable to estimation using least squares or standard maximum likelihood techniques. To estimate the model therefore, we propose and implement maximum simulated likelihood with importance sampling. The results suggest a mutual dependence (feedback) between technical efficiency and delays. Policy implications are derived. [ABSTRACT FROM AUTHOR]