CHAPTER

William H. Greene

in The Measurement of Productive Efficiency and Productivity Change

Published in print February 2008 | ISBN: 9780195183528 
Published online January 2008 | e-ISBN:
9780199870288 | DOI: http://dx.doi.org/10.1093/acprof:oso/9780195183528.003.0002

PREVIEW
This chapter provides an overview of techniques for econometric analysis of technical (production) and
economic (cost) efficiency. The stochastic frontier model of Aigner, Lovell, and Schmidt (1977) is now the
standard econometric platform for this type of analysis. The underlying models and econometric
techniques that have been used in studying technical inefficiency in the stochastic frontier framework are
surveyed, and some of the recent developments in econometric methodology are presented. Applications
that illustrate some of the computations are presented in the final section.

Keywords: econometric analysis; production efficiency; econometric analysis; cost efficiency; stochastic

frontier model

Chapter.  72320 words.  Illustrated.

Subjects: Econometrics and Mathematical Economics

Data envelopment analysis (DEA) is a nonparametric method in operations research and

economics for the estimation of production frontiers [clarification needed]. It is used to empirically
measure productive efficiency of decision making units (or DMUs).

Stochastic frontier analysis (SFA) is a method of economic modeling. It has its starting point in
the stochastic production frontier models simultaneously introduced by Aigner, Lovell and Schmidt
(1977) and Meeusen and Van denk Broeck (1977).

The production frontier model without random component can be written as:

 the best where yi is the observed scalar output of the producer i, i=1,..I, xi is a vector

of N inputs used by the producer i, f(xi, β) is the production frontier, and   is a vector of
technology parameters to be estimated.
TEi denotes the technical efficiency defined as the ratio of observed output to maximum feasible
output. TEi = 1 shows that the i-th firm obtains the maximum feasible output, while TEi < 1 provides
a measure of the shortfall of the observed output from maximum feasible output.

A stochastic component that describes random shocks affecting the production process is added.
These shocks are not directly attributable to the producer or the underlying technology. These
shocks may come from weather changes, economic adversities or plain luck. We denote these

effects with  . Each producer is facing a different shock, but we assume the shocks are
random and they are described by a common distribution

The second edition of this book has been written for the same audience as the first edition. It is designed
to be a "first port of call" for people wishing to study efficiency and productivity analysis. The book
provides an accessible introduction to the four principal methods involved: econometric estimation of
average response models; index numbers; data envelopment analysis (DEA); and stochastic firontier
analysis (SFA). For each method, we provide a detailed introduction to the basic concepts, give some
simple numerical examples, discuss some of the more important extensions to the basic methods, and
provide references for further reading. In addition, we provide a number of detailed empirical applications
using real-world data. The book can be used as a textbook or as a reference text. As a textbook, it
probably contains too much material to cover in a single semester, so most instructors will want to design
a course around a subset of chapters. For example, Chapter 2 is devoted to a review of production
economics and could probably be skipped in a course for graduate economics majors. However, it should
prove useful to undergraduate students and those doing a major in another field, such as business
management or health studies

Stochastic frontier analysis

From Wikipedia, the free encyclopedia

Stochastic frontier analysis (SFA) is a method of economic modeling. It has its starting point in
the stochastic production frontier models simultaneously introduced by Aigner, Lovell and Schmidt
(1977) and Meeusen and Van denk Broeck (1977).
The production frontier model without random component can be written as:

 the best where yi is the observed scalar output of the producer i, i=1,..I, xi is a vector

of N inputs used by the producer i, f(xi, β) is the production frontier, and   is a vector of
technology parameters to be estimated.
TEi denotes the technical efficiency defined as the ratio of observed output to maximum feasible
output. TEi = 1 shows that the i-th firm obtains the maximum feasible output, while TEi < 1 provides
a measure of the shortfall of the observed output from maximum feasible output.
A stochastic component that describes random shocks affecting the production process is added.
These shocks are not directly attributable to the producer or the underlying technology. These
shocks may come from weather changes, economic adversities or plain luck. We denote these

effects with  . Each producer is facing a different shock, but we assume the shocks are
random and they are described by a common distribution.
The stochastic production frontier will become:

We assume that TEi is also a stochastic variable, with a specific distribution function, common to all
producers.

We can also write it as an exponential  , where ui ≥ 0, since we required TEi ≤ 1. Thus, we

obtain the following equation:

Now, if we also assume that f(xi, β) takes the log-linear Cobb-Douglas form, the model can be written
as:

where vi is the “noise” component, which we will almost always consider as a two-sided normally
distributed variable, and ui is the non-negative technical inefficiency component. Together they
constitute a compound error term, with a specific distribution to be determined, hence the name of
“composed error model” as is often referred.
Stochastic Frontier Analysis has examined also "cost" and "profit" efficiency (see Kumbhakar &
Lovel 2003). The "Cost frontier" approach attempts to measure how far from full-cost minimization
(i.e. cost-efficiency) is the firm. Modeling-wise, the non-negative cost-inefficiency component is
added rather than subtracted in the stochastic specification. "Profit frontier analysis" examines the
case where producers are treated as profit-maximizers (both output and inputs should be decided by
the firm) and not as cost-minimizers, (where level of output is considered as exogenously given).
The specification here is similar with the "production frontier" one.
Stochastic Frontier Analysis has also been applied in micro data of consumer demand in an attempt
to benchmark consumption and segment consumers. In a two-stage approach, a stochastic frontier
model is estimated and subsequently deviations from the frontier are regressed on consumer
characteristics (Baltas 2005)
Data envelopment analysis (DEA) is a nonparametric method in operations
research and economics for the estimation of production frontiers [clarification needed]. It is used to empirically
measure productive efficiency of decision making units (or DMUs). Although DEA has a strong link
to production theory in economics, the tool is also used for benchmarking in operations
management, where a set of measures is selected to benchmark the performance of manufacturing
and service operations. In the circumstance of benchmarking, the efficient DMUs, as defined by
DEA, may not necessarily form a “production frontier”, but rather lead to a “best-practice frontier”
(Cook, Tone and Zhu, 2014). DEA is referred to as "balanced benchmarking" by Sherman and Zhu
(2013).Non-parametric approaches have the benefit of not assuming a particular functional
form/shape for the frontier, however they do not provide a general relationship (equation) relating
output and input. There are also parametric approaches which are used for the estimation of
production frontiers (see Lovell & Schmidt 1988 for an early survey). These require that the shape of
the frontier be guessed beforehand by specifying a particular function relating output to input. One
can also combine the relative strengths from each of these approaches in a hybrid method (Tofallis,
2001) where the frontier units are first identified by DEA and then a smooth surface is fitted to these.
This allows a best-practice relationship between multiple outputs and multiple inputs to be estimated.

"The framework has been adapted from multi-input, multi-output production functions and applied in
many industries. DEA develops a function whose form is determined by the most efficient producers.
This method differs from the Ordinary Least Squares (OLS) statistical technique that bases
comparisons relative to an average producer. Like Stochastic Frontier Analysis (SFA), DEA identifies
a "frontier" which are characterized as an extreme point method that assumes that if a firm can
produce a certain level of output utilizing specific input levels, another firm of equal scale should be
capable of doing the same. The most efficient producers can form a 'composite producer', allowing
the computation of an efficient solution for every level of input or output. Where there is no actual
corresponding firm, 'virtual producers' are identified to make comparisons" (Berg 2010).

Attempts to synthesize DEA and SFA, improving upon their drawbacks, were also made in the
literature, via proposing various versions of non-parametric SFA [1]and Stochastic DEA.[2]