Title: From the analysis of the composite indicators to the analysis of the symbolic composite indicators
Authors: Carlo Drago - University of Rome Niccolo Cusano (Italy) [presenting]
Abstract: Nowadays, composite indicators represent and summarize complex phenomena that cannot be measured by looking at a single variable. In this sense, the advantage of using composite indicators is that they can be used directly in policy analysis. However, composite indicators are based on assumptions that may vary and are subjective (the treatment of missing data, weighting, the choice of variables to use, etc.). In this context, we propose using symbolic composite indicators, which are more informative than the original composite indicators since they explicitly internalize or take into account the variability of the original composite indicators when considering the different relevant choices on the factors. In this sense, sensitivity analysis is directly incorporated into the construction of the indicator. It could be considered possible to interpret the parameters of the final indicators.