Partenaires

CNRS
Logo tutelle
Logo tutelle
Logo tutelle


Rechercher

Sur ce site

Sur le Web du CNRS


Accueil du site > Séminaires > Probabilités Statistiques et réseaux de neurones > Spatial Prediction for Massive Datasets

Vendredi 16 mars 2007 à 11h00

Spatial Prediction for Massive Datasets

Noel Cressie (Ohio State University, USA), ncressie@stat.ohio-state.edu

Résumé : Environmental datasets obtained from satellites are typically massive in size. The massiveness causes problems in computing optimal spatial (kriging) predictors. In this talk, a flexible family of nonstationary covariance functions is constructed using a set of basis functions that is fixed in number. This results in computational simplications in deriving the kriging predictor and its kriging variance. We call the methodology fixed rank kriging (FRK) and we apply it to a large dataset of remotely sensed Total Column Ozone (TCO) data, observed over the entire globe. This talk represents joint research with Gardar Johannesson.

Dans la même rubrique :