High Dimensional & Complex Data: functional, spatial and object data

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 or satellite remote sensing, system control datasets, 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, fuelling many fascinating and fast growing research areas of modern statistics.

In particular the group is involved in the study of:

Mobile Phone Treelet Analysis

  • functional data: smoothing, alignment and registration, exploratory data tools, regression models, parametric and non-parametric inferential tools, depth measures, methods for multidimensional functional data, methods for spatially and temporally dependent functional data, methods for functional compositional data;
  • spatial and space-time data: geostatistical models and methods for spatially and temporally dependent functional data and for object data (see more on CompGeo), spatial regression models with differential regularizations (more), methods for data distributed over complex domains and over manifold domains (more);
  • non-Euclidean data and object data;
  • omics data: methods for the analysis of Next Generation Sequencing data;
  • blind-source separation;
  • reputational risk assessment.




Case studies:Spatial Regression Brain

  • the AneuRisk project;
  • analysis of particle-size distributions at the Lauswiesen site (Germany) (see more on CompGeo);
  • analysis of ECG signals of PROMETEO project (see more on BioMath).