Abstract
Composite indicators (CI) are widely employed to synthesize complex multidimensional phenomena, providing a framework for relative comparisons across units. Traditional frontier-based approaches, such as the Benefit of the Doubt model (BoD), often assume compensability among indicators and do not adequately account for contextual and geographic heterogeneity. These limitations can result in biased benchmarks and reduce the interpretability of results. Among other approaches proposed in the literature, our research has advanced the field by integrating non-compensatory structures, robust frontier techniques, and spatial extensions. This combination allows performance to be disentangled from contextual influences, enhancing both the accuracy and relevance of assessments of the studied phenomenon. By reducing arbitrariness in weighting and aggregation, these methodological contributions improve the transparency, robustness, and policy relevance of composite indicators.
Link Microsoft Teams
Organizations:
Silvia Emili