B1756
Title: A multivariate permutation test for the analysis of C paired samples in the presence of multiple data types
Authors: Rosa Arboretti - University of Padova (Italy)
Elena Barzizza - Università degli studi di Padova (Italy)
Riccardo Ceccato - University of Padova (Italy) [presenting]
Luigi Salmaso - University of Padova (Italy)
Abstract: The Nonparametric Combination (NPC) is a flexible permutation-based methodology that can be adopted to deal with a wide range of complex problems, including the comparison of two or more populations when a multivariate outcome is observed. We propose on a new NPC-based testing procedure to address a specific multivariate problem in which C 2 paired samples and multiple data types are available. A simulation study is proposed to evaluate the performances of our proposal under several challenging scenarios. A real-data application is also considered. Data were gathered through a questionnaire that was submitted to multiple respondents, asking them to evaluate a product in terms of a certain set of KPIs after multiple time frames. A number of experiments were also conducted and several continuous KPIs were measured after the same time frames. The NPC-based test was therefore adopted to compare the performances of the product across time.
