Comparer des méthodes

	Krigage universel bayésien	Autocorrélation spatiale
Domaine	Analyse spatiale	Analyse spatiale
Famille	Regression model	Regression model
Année d'origine	1990s–2000s	1950
Auteur d'origine	Diggle, Tawn & Moyeed; Kitanidis; Handcock & Stein	P. A. P. Moran (global measure, 1950); Roy Geary (Geary's C, 1954); Luc Anselin (LISA, 1995)
Type	Bayesian geostatistical interpolation with trend	Spatial statistic / exploratory spatial data analysis
Source fondatrice	Diggle, P. J., & Ribeiro, P. J. (2007). Model-Based Geostatistics. Springer. ISBN: 978-0387329079	Moran, P. A. P. (1950). Notes on continuous stochastic phenomena. Biometrika, 37(1/2), 17–23.
Alias	BUK, Bayesian kriging with trend, Bayesian spatial interpolation with covariates, stochastic universal kriging	spatial dependence, geographic autocorrelation, spatial clustering measure, SA
Apparentées	6	5
Résumé	Bayesian Universal Kriging (BUK) extends classical universal kriging by placing prior distributions on trend coefficients and spatial covariance parameters, then propagating full posterior uncertainty into predictions. It interpolates spatially referenced continuous data while simultaneously estimating large-scale deterministic trends driven by covariates and small-scale stochastic spatial dependence, yielding prediction intervals that honestly account for both parameter and interpolation uncertainty.	Spatial autocorrelation quantifies the degree to which a variable's values at nearby locations resemble each other more (positive autocorrelation) or less (negative autocorrelation) than expected by chance. Global indices such as Moran's I summarise the pattern across the entire study area, while local variants reveal clusters and outliers at the level of individual observations.