The statistical analysis of complex objects is the subject of study of Object Oriented Data Analysis. The investigation of this area involves the formulation of new statistical tools as generalization of the classical ones. In this talk we consider network-valued data: the data points of the population under scrutiny are networks. We propose a statistical framework to compare two samples of networks within a permutational approach. The method is tested via simulations, and an application to real data concerning the use of the bike sharing service in Milan is presented.
Lovato, I., Pini, A., Stamm, A., Vantini, S., Non-parametric inference for network-valued data, Paper, in Cladag 2017 Meeting of the Classification and Data Analysis Group Book of Short Papers, (Milano, 13-15 September 2017), Universitas Studiorum S.r.l. Casa Editrice, Mantova 2017: 1-4 [http://hdl.handle.net/10807/119854]
Non-parametric inference for network-valued data
Pini, A.;
2017
Abstract
The statistical analysis of complex objects is the subject of study of Object Oriented Data Analysis. The investigation of this area involves the formulation of new statistical tools as generalization of the classical ones. In this talk we consider network-valued data: the data points of the population under scrutiny are networks. We propose a statistical framework to compare two samples of networks within a permutational approach. The method is tested via simulations, and an application to real data concerning the use of the bike sharing service in Milan is presented.
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
