Cluster Analysis of Transactional Data in the Frequency Domain
Ivan Luciano Danesi, Flavio Pons, Cristina Rea · 48th Scientific Meeting of the Italian Statistical Society · 2016
We consider the problem of clusterization and discrimination of stochastic processes with an application to transactional data. The analysed dataset is a collection of payment card transactions, referred to a selected number of Point of Sales (opportunely discretized). The aim is to identify homogeneous groups of Point of Sales starting by the received payments. We use an approach based on the variance profile. Estimation is carried out in the frequency