Data Envelopment Analysis For Performance Measurement

Data Envelopment Analysis for
Performance Measurement
Introduction
In evaluating the performance of a business the owners or the managers
would typically like to know:
• Is the company making best use of the resources?

• Is it possible to produce more from the same input bundle? If so, which
outputs and how much more?
• Can the firm economize on the resources used? If so, which inputs and by
how much?
• Is the firm's input-mix consistent with the relative prices of the inputs? If
not, which inputs should be substituted for hat?
• Is the firm of the right size? If not, is it too big or too small?
• Would a potential merger with another specific firm enhance efficiency?
The list goes on.

In order to evaluate the performance of any decision-making unit -
• Needs to define a best performance
• Two main approaches available:

• Stochastic Frontier Analysis (SFA): parametric and econometric
approach
• Data Envelopment Analysis (DEA) that uses mathematical
programming techniques.
Effectiveness, Productivity, and Efficiency
Total factor productivity (TFP)
A natural solution would be to take some average of the partial productivities
Define its total factor productivity as the weighted geometric mean
Tornqvist index
Now take a closer look at the input aggregator function
When the inputs L and K are aggregated through a production function, the
total factor productivity of a firm also becomes its technical efficiency index.
The production function defines the maximum quantity of output that can be
produced from a given input bundle. Thus,
Stochastic Frontier Models
• Motivation:
• Factors not under control of the firm
• Measurement error
• Differential rates of adoption of technology
• Frontier is randomly placed by the whole collection of
stochastic elements which might enter the model outside the
control of the firm.
• Aigner, Lovell, Schmidt (1977), Meeusen, van den Broeck
(1977), Battese, Corra (1977)
The Stochastic Frontier Model
yi = f (xi )TE i e v i
ln y i =  +   xi + vi  ui
=  +  xi +  i .

ui > 0, but vi may take any value. A symmetric

distribution, such as the normal distribution, is usually
assumed for vi. Thus, the stochastic frontier is
+’xi+vi
and, as before, ui represents the inefficiency.

DEA
DEA requires no parametric specification of the production frontier
and relies on a number of fairly general assumptions about the
nature of the underlying production technology. Using a sample of
actually observed input-output data and these assumptions, it
derives benchmark output quantity with which the actual output of
a firm can be compared for efficiency measurement.
Assumptions About the Technology
1. All actually observed input-output bundles are feasible.
2. The production possibility set is convex.
3. Inputs are freely disposable.
4. Outputs are freely disposable

Technical Efficiency: Output oriented measure
Technical Efficiency: Input oriented measure
A one-input one-output example: Geometry of DEA
Two input one output example
Technology Sets, e.g., GR and DEA
An Algebraic Formulation of the DE A Optimization Problem
DEA Linear Programming for input Technical Efficiency
Technical Efficiency
Returns to Scale and Technical Efficiency
• The DEA problems we have specified so far to estimate Farrell input
and output based technical efficiency have restricted technology to
satisfy:
• constant returns to scale
• strong disposability of inputs and outputs
• The returns to scale of technology are determined by the restrictions
on the intensity variables, i.e., the z’s.
• The disposability property arises from the inequalities on the input
and output constraints.
DEA and Returns to Scale
Non-increasing Returns to Scale and DEA
Variable Returns to Scale and DEA
Scale Efficiency
Input Based Scale Efficiency
Output Based Scale Efficiency
Cost Minimisation and Decomposition of Efficiency
Productivity Growth
y/x
Malmquist Productivity index
THE MALMQUIST OUTPUT-BASED INDEX OF TOTAL FACTOR PRODUCTIVITY AND
OUTPUT DISTANCE FUNCTIONS
DEA Summary
• Addresses fundamental productivity measurement problems
due to ...
• complexity of service outputs
• variability in service outputs
• Provides useful information
• objective measures of productivity
• reference set of comparable units
• excess use of inputs measure
• returns to scale measure
DEA Summary (cont.)
• Role of DEA
• “data mining” to generate hypotheses
• evaluation/measurement
• benchmarking to identify “best practice” units
• Caveats
• “black box” - No information on root causes of inefficiency
• Be aware of assumptions (e.g. linearity)
• Can be sensitive to selection of inputs/outputs
Thank you !

Data Envelopment Analysis For Performance Measurement

Uploaded by

Copyright:

Available Formats

Data Envelopment Analysis For Performance Measurement

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Data Envelopment Analysis For Performance Measurement

Uploaded by

Copyright:

Available Formats

Data Envelopment Analysis for

• Is the company making best use of the resources?

The list goes on.

• Needs to define a best performance

• Two main approaches available:

A natural solution would be to take some average of the partial productivities

Define its total factor productivity as the weighted geometric mean

ui > 0, but vi may take any value. A symmetric

and, as before, ui represents the inefficiency.

1. All actually observed input-output bundles are feasible.

2. The production possibility set is convex.

3. Inputs are freely disposable.

4. Outputs are freely disposable

You might also like

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.