Advantage Parametric Methods of Efficiency Analysis Parametric methods of efficiency analysis have significant ad
-
vantages by providing the possibilities to use panel data, to distin-
guish the random noise from inefficiency and to calculate the stan
-
dard error of efficiency measurement results [11]. In addition, the
primary advantage of the parametric methods of SFA, DFA and TFA
lies in their ability to allow for random error in efficiency estima
-
tion [6]. Similarly, this approach is a flexible technique in measuring
the frontier production function, which provides for a meaningful
estimate of the measurement error [17]. Therefore, parametric
methods of efficiency have the benefit of modeling inefficiency
and noise [1]. The SFA method is preferable when certain classi-
cal assumptions are satisfied regarding the composite error terms,
including the contributions from the inefficiency distribution and
measurement errors [8].
In addition, this method of efficiency analysis is useful to mea
-
sure quantitatively independent of the effect of other exogenous
factors such as market prices, through exploiting a programming or
econometric method to control for the effects [6]. Moreover, Ajibe-
fun [2] reported that parametric frontier analysis allows the test
of hypothesis concerning the goodness of fit of the model. On the
other hand, parametric frontier functions require the definition of a
specific functional form for the technology and for the inefficiency
error term [4].
Disadvantage Parametric Methods of Efficiency Analysis The major disadvantage of these methods of efficiency analysis
is that it requires specification of the technology, which may be re
-
strictive in most cases [2]. In addition, conducting the parametric
approaches is how to appropriately distinguish random noise from
true in efficiency, as neither of them is observable [6]. The func
-
tional form requirement causes both specification and estimation
problems [4].
Furthermore; parametric method of efficiency analysis doesn’t
impose axiomatic properties in the estimate the frontier [1]. In SFA
studies, an assumption regarding to a specific functional form of
stochastic frontier is required a priori and wrong choice of produc-
tion function may influence the results. Moreover, the maximum
likelihood does not allow assessing the reliability of inferences in
small samples. Therefore, SFA requires using of large number of
DMUs [10].