Statistical Analysis of High Dimensional & Complex Data

Featured, MOX. Dept. Mathematics, Politecnico Milano

The last years have seen the advent and development of many devices able to record and store an always increasing amount of complex and high dimensional data: 3D images generated by medical scanners, satellite remote sensing, system control datasets, administrative databases, mobile phone data, etc. The analysis of these complex and high dimensional data poses new challenging problems and requires the development of novel statistical models and computational methods, fueling many fascinating and fast-growing research areas of modern statistics. It provides also an invaluable insight for the control and optimization of key processes of business and industrial sectors, of life-sciences, and of healthcare management.

The Applied Statistics group at MOX, in collaboration with domain experts from the private and public sector, is involved in the study of: functional data, compositional data, tensor data, neuroimaging and image data, network data, spatial and space-time data, epidemiological, healthcare, administrative, pharmaceutical data, and vital signals.

One-day-ahead prediction of supply and demand curves in the Italian natural gas market (A. Canale, and S. Vantini (2016): “Constrained functional time series: Applications to the Italian gas market”, International Journal of Forecasting, Vol. 32(4), pages 1340-1351).

 

Spatial prediction of particle size densities in an experimental site in Tübingen, Germany (A. Menafoglio, P. Secchi (2017): “Statistical analysis of complex and spatially dependent data: a review of Object Oriented Spatial Statistics”, European Journal of Operational Research, 258(2), pages 401–410).

 

Quality control of foamed material production (A. Menafoglio, M. Grasso, P. Secchi, B.M. Colosimo, (2018): “Profile Monitoring of Probability Density Functions via Simplicial Functional PCA with application to Image Data”, Technometrics, forthcoming).

 

Remote monitoring of laser welding processes (A. Pini, S. Vantini, B. M. Colosimo, M. Grasso (2018): “Domain-selective functional analysis of variance for supervised statistical profile monitoring of signal data”. Journal of the Royal Statistical Society C, Vol. 67, pages 55–81).

 

Estimation of population dynamics in urban areas (P. Secchi, S. Vantini, V. Vitelli (2015): “Analysis of spatio-temporal mobile phone data: a case study in the metropolitan area of Milan ” [with discussion], Statistical Methods and Applications, Vol. 24(2), pages 279-300).