[Statlist] Research Seminar in Statistics *FRIDAY 29 NOVEMBER 2019* GSEM, University of Geneva

[gsem-support-instituts]

Dear All,

We are pleased to invite you to our next Research Seminar.

Looking forward to seeing you


Organizers :                                                                                   
E. Cantoni - S. Engelke - D. La Vecchia - E. Ronchetti
S. Sperlich - F. Trojani - M.-P. Victoria-Feser


FRIDAY, 29 NOVEMBER 2019 at 11:15am, Uni-Mail M 5220

Interpolation for Huge Spatial Datasets: Theory, Implementations and Ideas
Reinhard FURRER - University of Zurich

ABSTRACT:
"The need for computationally efficient and easy-to-use methodologies to interpolate huge spatial datasets is undisputed. However, classical kriging methodologies and corresponding software do not scale well due to the inherent limit of the covariance matrix that can be handled. On the other hand ignoring their inherent dependency structures and/or nonstationarities results in incorrect uncertainty estimates and may lead to wrong conclusions.

In this talk, I discuss some recent theoretical results for huge spatial datasets for which the implementation has essentially no memory limitations (at the price of computing time).  More specifically, we study estimation and prediction of Gaussian random fields with covariance models belonging to the generalized Wendland (GW) class, under fixed domain asymptotics. We characterize the equivalence of two Gaussian measures with GW covariance function, and we provide sufficient conditions for the equivalence of two Gaussian measures with Matérn and GW covariance functions."


Visit the website: https://www.unige.ch/gsem/en/research/seminars/rcs/