Hierarchical Modeling and Analysis for Spatial Data

By T. Coburn. 2007

592 citations recorded by the source corpus.

Subfields: Geography, Planning and Development, Signal Processing

View Paper: https://www.semanticscholar.org/paper/687c72ef5706b38a92c377464e95c8add8656390

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Geography, Planning and Development

In Signal Processing

Citations in the corpus

The relative efficiencies of various predictors in spatial econometric models containing spatial lags (2007, methodology): “Kelejian and Prucha 11Also note that (9), (18) and (22) imply that MSE(y (3) i ) = var(εi|y−i) ≤ var(εi) =MSE(y (5) i ).”
Bayesian Spatial Modeling of Extreme Precipitation Return Levels (2007, methodology): “Berger, DeOiveira, and Sanso (2001) and Banerjee et al. (2004) cautioned that improper priors for the sill and range parameters generally result in improper posteriors.”
Estimation of a non-stationary model for annual precipitation in southern Norway using replicates of the spatial field (2014, methodology): “Gaussian random fields (GRFs) play an important role in spatial statistics; they are commonly used to model the spatial process itself, or as building blocks in hierarchical models (see e.g. Banerjee et al., 2004; Cressie and Wikle, 2011).”
Gaussian predictive process models for large spatial data sets (2008, methodology): “With hierarchical models, we assign prior distributions to Ω = .β, θ, τ 2 / and inference proceeds by sampling from p.Ω|Y/, whereas prediction at an arbitrary site s 0 samples p{Y.s 0 /|Y} one for one with posterior draws of Ω by composition (Banerjee et al., 2004).”
On the relationship between conditional (CAR) and simultaneous (SAR) autoregressive models (2017, methodology): “…for data collected at spatial locations into two broad classes: 1) geostatistical models with continuous spatial support, and 2) lattice models, also called areal models (Banerjee et al., 2004), where data occur on a (possibly irregular) grid, or lattice, with a countable set of nodes or locations.”
Spatial designs and properties of spatial correlation: Effects on covariance estimation (2007, methodology): “While we have focused on ML and REML estimation, a comment regarding Bayesian estimation is warranted (Banerjee et al. 2004, pp. 129–170).”
Programming With Models: Writing Statistical Algorithms for General Model Structures With NIMBLE (2015, background): “…non-independence among data – from spatial, temporal, clustered or other sources of shared variation – that are often modeled using unobserved (often unobservable) random variables in a hierarchical model structure (e.g., Banerjee et al., 2003; Royle and Dorazio, 2008; Cressie and Wikle, 2011).”
Spatial and spatio-temporal models with R-INLA. (2013, methodology): “In fact, spatial GFs are aﬀected by the so called “big n problem” (Jona Lasinio et al., 2012; Banerjee et al., 2004), which is due to the computational costs of O ( n 3 ) to perform matrix algebra operations with n × n dense covariance matrices (whose dimension is given by the number of…”
Geostatistics of Dependent and Asymptotically Independent Extremes (2013, background): “Classical geostatistics is a well-developed field surveyed in numerous textbooks (Cressie 1993; Wackernagel 2003; Banerjee et al. 2004; Diggle and Ribeiro 2007; Cressie and Wikle 2011), with much software available and a wide range of user communities corresponding to its many applications.”
Asymptotic properties of computationally efficient alternative estimators for a class of multivariate normal models (2007, methodology): “In this paper, we study the asymptotic properties of the estimators that minimize three alternatives to the likelihood function, designed to increase the computational efficiency.”
A Case Study Competition Among Methods for Analyzing Large Spatial Data (2017, unknown): “For decades, the Gaussian process (GP) has been the primary tool used for the analysis of geostatistical (point-referenced) spatial data (Schabenberger and Gotway 2004; Cressie 1993; Cressie andWikle 2015; Banerjee et al. 2014).”
Spatio-temporal modeling of particulate matter concentration through the SPDE approach (2012, background): “Even if a GF is easily deﬁned directly through its ﬁrst and second moments, its implementation suﬀers from the so-called “big n problem” (Banerjee et al 2004, page 387), that arises especially in case of large datasets in space and time.”
Spatial Analysis to Quantify Numerical Model Bias and Dependence (2008, background): “(See Banerjee, Carlin, and Gelfand 2004 for a hierarchical Bayesian modeling approach for nonnested block-level realignment.)”