Intelligent synthetic composite indicators with application

This article is proposing an alternative approach to develop Intelligent Synthetic Composite Indicators (ISCI). The suggested approach utilizes Fuzzy Proximity Knowledge Mining technique to build the qualitative taxonomy initially, and then Fuzzy c-means is employed to form the new composite indicators. A fully worked application is presented. The application uses Information and Communication Technology real variables to form a new unified ICT index, illustrating the method of construction for ISCI. The weighting and aggregation results obtained were compared against Principal Component Analysis, Factor Analysis and the Geometric mean to weight and aggregate synthetic composite indicators. This study also compares and contrasts two special Fuzzy c-means techniques that is, the Optimal Completion Strategy and the Nearest Prototype Strategy to impute missing values. The results are compared against statistical imputation techniques. The validity and robustness of all techniques are evaluated using Monte Carlo simulation. The results obtained suggest a novel, intelligent and non-biased method of building future composite indicators.