SHAPE - Statistical Models For Human Perception And Evaluation

The main aim of the Scientific project was to unify several approaches for modelling ordinal data by exploiting disseminated competences on alternative methodologies and performing an international experience on well documented data sets of real importance in different fields (especially in Sensory sciences, Medicine, Economics, Psychology). Thus, the project improved the decisional process by adding a more consistent and flexible approach to the current state-of-the-art. A focal point has been the search for similarities and differences among the statistical methods and the implementation of new paradigm with some extensions to multivariate perspectives. More specifically, the topics of the project involved about forty scholars (from Italy and abroad) in different scientific fields who contributed with expertise, workshops, conferences and papers. National and international conferences have been organized and 33 papers have been published in statistical journals. Then, more than 10 papers are currently under revision. A package in the R environment for the implementation of the mixture models CUB has been delivered. A post-doc researcher has been recruited so as to be trained on the methodological issues of the SHAPE project.