Abstract
Xu and Ouenniche (2012a) proposed an input-oriented radial super-efficiency Data Envelopment Analysis (DEA) based model to address a common methodological issue in the evaluation of competing forecasting models; namely, ranking models based on a single performance measure at a time, which typically leads to conflicting ranks. However, their approach suffers from a number of issues. In this paper, we overcome these issues by proposing a slacks-based context-dependent DEA framework and use it to rank forecasting models of oil prices volatility.
Highlights
T he design of quantitative models for forecasting continuous variables in a wide range of application areas has attracted the attention of a large number of academics and professionals for some time; the performance evaluation of competing forecasting models has not received as much attention
Notice that different criteria led to different unidimensional rankings, which provides evidence of the problem resulting from the use of a unidimensional approach in a multicriteria setting
For our data set – see Table 4, the efficient model SMA20 maintained its best position in the rankings regardless of the type of Data Envelopment Analysis (DEA) analysis, because it is always on the best efficient frontier and has zero slacks regardless of the performance measures used
Summary
T he design of quantitative models for forecasting continuous variables in a wide range of application areas has attracted the attention of a large number of academics and professionals for some time; the performance evaluation of competing forecasting models has not received as much attention. A super-efficiency data envelopment analysis model has been proposed by Xu and Ouenniche (2011) to devise a multi-criteria ranking of competing forecasting models of oil prices’ volatility; their approach suffers from the following issues. Radial DEA models could only take account of technical efficiency and ignore potential slacks in inputs and outputs and may over-estimate efficiency scores. Within a super-efficiency DEA framework, super-efficiency scores are used to rank order the efficient DMUs; the reference set changes from one efficient DMU evaluation to another, which in some contexts might be viewed as “unfair” benchmarking. We overcome these issues by proposing a slacks-based context-dependent DEA (CDEA) framework (Morita, Hirokawa, & Zhu, 2005; Seiford & Zhu, 2003) for assessing the relative performance of competing volatility forecasting models.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
