Network inference from high-dimensional data: on covariates, missingness and normality

Abstract: Statistical inference of networks using graphical modelling approaches has applications in many fields, such as biology, finance and health. The resulting data have often high levels of complexity and dimensionality. This talk will highlight recent advances in the development of computationally efficient approaches for network inference in these non-standard settings, both in a frequentist and Bayesian paradigm. We will consider closely the case of heterogeneous, missing and nonGaussian data, with examples taken from genomics.

Eventi passati

È possibile consultare gli eventi del precedente ciclo alla pagina https://webmagazine.unitn.it/evento/dmath/67573/maths-bites-trento