Comparer des méthodes

	Co-Krigage Bayésien	Régression Spatiale Bayésienne
Domaine	Analyse spatiale	Analyse spatiale
Famille	Regression model	Regression model
Année d'origine	1990s–2000s	1990s–2000s
Auteur d'origine	Gelfand, Banerjee & colleagues; building on Matheron's cokriging framework	Banerjee, Carlin & Gelfand (foundational treatment); building on Besag (1974) for lattice priors
Type	Bayesian spatial interpolation	Bayesian hierarchical regression
Source fondatrice	Diggle, P. J., & Ribeiro, P. J. (2007). Model-Based Geostatistics. Springer. ISBN: 978-0387329079	Banerjee, S., Carlin, B. P., & Gelfand, A. E. (2015). Hierarchical Modeling and Analysis for Spatial Data (2nd ed.). CRC Press. ISBN: 978-1439819173
Alias	Bayesian cokriging, Bayesian co-regionalization, BCK, Bayesian multivariate kriging	Bayesian hierarchical spatial model, BSR, Bayesian geostatistical regression, Bayesian spatial linear model
Apparentées	5	3
Résumé	Bayesian Co-Kriging is a multivariate geostatistical method that uses auxiliary spatially correlated variables to improve predictions of a primary variable of interest. By placing Bayesian priors on cross-covariance parameters, it propagates all uncertainty — including parameter uncertainty — into the prediction intervals, yielding fully probabilistic maps with calibrated uncertainty bounds.	Bayesian Spatial Regression embeds a spatially structured random effect into a regression framework and estimates all parameters — including spatial range and variance — through posterior inference rather than point estimation. It handles spatial autocorrelation, quantifies full predictive uncertainty, and accommodates small or irregular spatial datasets via hierarchical priors.