Abstract
An extension of the translog cost function, where some trigonometric terms are added to gain better fit to the data, has gained popularity in recent literature examining efficiency rankings of banks. This paper demonstrates that the goodness-of-fit criterion is an unreliable indicator of forecasting ability, thus casting doubts on the validity of conclusions derived from the use of above frontier cost functions.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have