This project aims to sustain, strengthen and create scientific collaborations in statistical modeling  and nonparametric inference for complex data, involving Chilean, French and Uruguayan. The research project is divided into four core axes:

• Data constrained in a bounded domain of R^d and statistical modeling of animal trajectories by hypoelliptic diffusions with boundary conditions.
• Compositional data, with a clinical application to microbiome analysis.
• Model selection for spatial weighted regression, a model of great importance for geostatistical applications.
• Statistical tests and nonparametric inference for the analysis of spatio-temporal data (such as telecommunication data) with a non-homogeneous Poisson process.