A hybrid framework for space-time modeling of environmental data

Auteur(s) et Affiliation(s)

LI, X.
School of Geography and Planning, Sun Yat-sen Univ., Guangzhou, Chine

Description :
The article presents a hybrid framework combining machine learning and statistical methods to address a modelisation of the nonlinearities and nonstationarities of environmental space-time series. A four-stage procedure is proposed and it is applied to forecast annual average temperature at 137 national meteorological stations in China. The hybrid framework achieves better forecasting accuracy.

Type de document :
Article de périodique

Source :
Geographical analysis, issn : 0016-7363, 2011, vol. 43, n°. 2, p. 188-210, nombre de pages : 23, Références bibliographiques : 28 ref.

Date :

Editeur :
Pays édition : Etats-Unis, Columbus, OH, Ohio State University Press

Langue :
Droits :
Tous droits réservés © Prodig - Bibliographie Géographique Internationale (BGI)