Mikko J. Sillanpää
Professor of Statistics, Research Unit Leader
Bayesian statistical methods
My research group considers developing new computationally efficient and practical statistical methods and their applications in biology, medicine and other fields. Bayesian analysis and data analytic methods using hierarchical models and Markov Chain Monte Carlo sampling methods are preferred in the group. In many of the problems we are working with, important parts of the solution are formed by the efficient handling of the big-data, use of data analytic tools having close connections to the machine learning theory, and algorithmic view of computational methods. Specifically, the present research interest address statistical variable selection methods having sparsity-inducing mechanism and different inferential methods for covariance/precision matrix and network structures.