[Statlist] Research Seminar in Statistics *FRIDAY 8 DECEMBER 2017* 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 - D. La Vecchia - E. Ronchetti
S. Sperlich - F. Trojani - M.-P. Victoria-Feser 

FRIDAY 8 DECEMBER 2017 at 11:15am, Uni-Mail M 5220

Modeling of non-Gaussian processes using Lévy indicator convolutions
Thomas OPITZ - BioSP, INRA, France

ABSTRACT:
Process convolutions yield random fields with flexible marginal distributions and dependence beyond Gaussianity, but frequentist statistical inference is often hampered by a lack of closed-form marginal distributions while simulation-based Bayesian inference may be prohibitively computer-intensive. We here remedy such issues through a class of process convolutions based on smoothing a (d+1)-dimensional Lévy basis with an indicator function kernel to construct a d-dimensional convolution process. Indicator kernels ensure univariate distributions in the Lévy basis family, which provides a sound basis for interpretation, parametric modeling and statistical estimation. We propose a class of stationary and isotropic convolution processes constructed through hypograph indicator sets defined as the space between the curve (s,H(s)) of a spherical probability density function H and the plane (s,0). If H is radially nonincreasing, the covariance is expressed through the univariate distribution function of H. The bivariate joint tail behavior in such convolution processes will be explored. We will further discuss modeling extensions beyond stationary and isotropic spatial models, including latent process constructions. For statistical inference of parametric models, we develop pairwise likelihood techniques and illustrate these on real data examples.


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