Creating Composite Indicators with DEA and Robustness Analysis: The Case of the Technology Achievement Index

Cherchye, Laurens; Moesen, Willem; Rogge, Nicky; Puyenbroeck, Tom Van; Saisana, Michaela; Saltelli, Andrea; Liška, Roman; Tarantola, Stefano

doi:10.2139/ssrn.888461

Cited by 76 publications

(105 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…In addition, it is unlikely that the (DSTI, 2003). Finally, in case of fixed weighting, country rankings may change by using another normalisation method (Cherchye et al, 2006). We conclude that equal weighting is a solution in case no other weighting method yields valid results.…”

Section: Equal Weightingmentioning

confidence: 87%

“…A general DEA model for indexes has been proposed in Cherchye et al (2006). Consider y ij as the raw value of indicator i (i = 1, .…”

Section: Data Envelopment Analysismentioning

confidence: 99%

“…Here, the contribution of each indicator pie share (i.e. the product of the indicator value and the indicator weight) to the composite indicator value is constrained to lie in the range defined by the minimum and maximum weight obtained from the budget allocation results over all experts (Cherchye et al, 2006). To get more robust restrictions we incorporate the opinion of four experts instead of two and take the average of the two highest weights (U m ) and the average of the two lowest weights (L m ) for each indicator.…”

Section: Data Envelopment Analysismentioning

confidence: 99%

“…Although the DEA domain is very extensive and numerous models exist, making the selection of an appropriate model a rather difficult task, this technique can be translated to be useful for composite indicators. This weighting method has already been used for a number of indexes (Cherchye et al, 2004(Cherchye et al, , 2006Cherchye and Kuosmanen, 2002) and to a limited extent in the road safety context (Odeck, 2005). Its strongest point is that the weights are endogenously determined and derived directly from the data (Cooper et al, 2000).…”

Section: Data Envelopment Analysismentioning

confidence: 99%

See 3 more Smart Citations

Combining road safety information in a performance index

Hermans

Bossche²,

Wets

2008

Accident Analysis & Prevention

159

104

View full text Add to dashboard Cite

a b s t r a c tIn this paper we focus on an essential step in the construction process of a composite road safety performance indicator: the assignment of weights to the individual indicators. In the composite indicator literature, this subject has been discussed for a long time, and no agreement has been reached so far. The aim of this research is to provide insights in the most important weighting methods: factor analysis, analytic hierarchy process, budget allocation, data envelopment analysis and equal weighting. We will give the essential theoretical considerations, apply the methods on road safety data from various countries and discuss their advantages and disadvantages. This will facilitate the selection of a justifiable method. It is shown that the position of a country in the ranking is influenced by the method used. The weighting methods agree more for countries with a relatively bad road safety performance. Of the five techniques, the weights based on data envelopment analysis resulted in the highest correlation with the road safety ranking of 21 European countries based on the number of traffic fatalities per million inhabitants. This method is valuable for the development of a road safety index.

show abstract

Section: Equal Weightingmentioning

confidence: 87%

“…A general DEA model for indexes has been proposed in Cherchye et al (2006). Consider y ij as the raw value of indicator i (i = 1, .…”

Section: Data Envelopment Analysismentioning

confidence: 99%

Section: Data Envelopment Analysismentioning

confidence: 99%

Section: Data Envelopment Analysismentioning

confidence: 99%

See 2 more Smart Citations

Combining road safety information in a performance index

Hermans

Bossche²,

Wets

2008

Accident Analysis & Prevention

159

104

View full text Add to dashboard Cite

show abstract

“…Since the first DEA-based CI model proposed by Melyn and Moesen (1991), various indexes have been developed by using the DEA technique. The environmental performance index [Färe et al, 2004], the human development index [Despotis, 2005], the macroeconomic performance index [Ramanathan, 2006], the sustainable energy index [Zhou et al, 2007], the internal market index [Cherchye et al, 2007], the technology achievement index [Cherchye et al, 2008], and the road safety performance index [Hermans, 2009], are examples among others. Furthermore, as a valuable extension of the basic DEA-based CI model, Shen et al (2011; proposed a generalized multiple layer DEA model (MLDEA) and a MLDEA-based CI model, which took the layered hierarchy of indicators into account.…”

Section: Methodsmentioning

confidence: 99%

Investigating Individual Driver Performance: Applying DEA on Simulator Data

Babaee

Shen

Hermans

et al. 2014

Advances in Intelligent Systems and Computing

View full text Add to dashboard Cite

The main purpose of the present study is to investigate individual driver's behavior by using the data from a driving simulator, in order to distinguish the best drivers and identify the problematic behavior of 'underperforming' drivers. To this end, 129 participants with different age and gender were enrolled to take part in a particular simulator scenario (i.e., curve taking) and their speed, acceleration and lateral position, the three most important driving performance indicators based on literature review, were monitored at various points (before, during and after the curve) while driving a STISIM simulator. As a widely accepted tool for performance monitoring, benchmarking and policy analysis, the concept of composite indicators (CIs), i.e., combining single indicators into one index score, was employed, and the technique of data envelopment analysis -an optimization model for measuring the relative performance of a set of decision making units, or drivers in this studywas used for the index construction. Based on the results from the model, all drivers were ranked, and valuable insight were gained by comparing the relative performance of each driver. Finally, the sensitivity of the results was examined.

show abstract

Settings and methods for global sensitivity analysis – a short guide

Saltelli

Saisana

2007

Proc Appl Math and Mech

View full text Add to dashboard Cite

Global sensitivity analysis offers a set of tools tailored to the impact assessment of certain assumptions on a models output. A recent book on the topic covers those issues [1]. Given the limited space for discussing thoroughly any of those methods, we will next summarize the main conclusions that derive from the application of various global sensitivity analysis methods on chemical models [2], econometric studies [3] financial models [4] and composite indicators [5,6].The incentive for subjecting a model to sensitivity analysis is linked to the aim of quantifying the relative importance of the model input factors and providing answers to specific questions, such as "Which of the uncertain input factor(s) is so non-influential that we can safely fix it/them?" or "If we could eliminate the uncertainty in one (or just few) of the input factors, which factor(s) should we choose, so as to reduce the most the variance of the output?" The choice of the proper sensitivity analysis technique depends on considerations such as: (a) computational cost of running the model, (b) number of input factors, (c) features of the model (e.g. linearity), (d) consideration of interactions among the input factors in the model, (e) setting for the analysis and audience. These characteristics may be used as criteria for deciding on the proper sensitivity analysis method to be used, although, as it is often the case, more than one method might be suitable for the same study.In case of models that require a modest amount of CPU time (i.e. up to the order of one minute per run), and with a number of input factors which does not exceed, for example, 20, the class of variance-based techniques (Sobol' method [7]) yields, convergence aside, the most complete and general pattern of sensitivity. An easy-to-code implementation [8] provides all the pairs of first-order and total effect sensitivity measures at a cost of (k + 1)N model runs, where is the number of factors and sample size N ≈ 500 ÷ 1000. The first order sensitivity measures capture the direct impact of an input factor, whilst the total effect sensitivity measures capture the direct and indirect (due to interactions) impact of an input factor. Any interaction term between two input factors can be easily computed at the additional cost of N model evaluations per sensitivity measure. The Sobol' method does not rely on any assumption about smoothness of the input-output mapping; it only relies on squareintegrability of the output variable Y . This is both an advantage and a limitation, since it implies a quite slow convergence rate of the estimator and therefore a computational cost that depends on the dimension k of the problem and on the relatively large N required for a reasonable accuracy of the sensitivity measures. In Saltelli's recipe [8], the sampling of the input factors is based on quasi-random numbers, which are sequences of multidimensional points characterized by the property of optimal space filling. In case of correlated input factors, an ad hoc computational scheme must ...

show abstract

Creating Composite Indicators with DEA and Robustness Analysis: The Case of the Technology Achievement Index

Cited by 76 publications

References 28 publications

Combining road safety information in a performance index

Combining road safety information in a performance index

Investigating Individual Driver Performance: Applying DEA on Simulator Data

Settings and methods for global sensitivity analysis – a short guide

Contact Info

Product

Resources

About