Asymptotic theory of generalized information criterion for geostatistical regression model selection

DOI10.1214/14-AOS1258MaRDI QIDQ482898zbMATH OpenOpenAlexFDO

Authors Chih-Hao Chang, Hsin-Cheng Huang, Ching-Kang Ing

Publication date 6 January 2015

Published in The Annals of Statistics (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1412.0836, https://projecteuclid.org/euclid.aos/1413810733

zbMATH Keywords

variable selectionselection consistencyAkaike's information criterionBayesian information criterionfixed domain asymptoticincreasing domain asymptotic

Mathematics Subject Classification ID

Statistical aspects of information-theoretic topics (62B10) Asymptotic properties of parametric estimators (62F12) Inference from spatial processes (62M30) Statistical ranking and selection procedures (62F07) Applications of statistics to environmental and related topics (62P12) Geostatistics (86A32)

Abstract: Information criteria, such as Akaike's information criterion and Bayesian information criterion are often applied in model selection. However, their asymptotic behaviors for selecting geostatistical regression models have not been well studied, particularly under the fixed domain asymptotic framework with more and more data observed in a bounded fixed region. In this article, we study the generalized information criterion (GIC) for selecting geostatistical regression models under a more general mixed domain asymptotic framework. Via uniform convergence developments of some statistics, we establish the selection consistency and the asymptotic loss efficiency of GIC under some regularity conditions, regardless of whether the covariance model is correctly or wrongly specified. We further provide specific examples with different types of explanatory variables that satisfy the conditions. For example, in some situations, GIC is selection consistent, even when some spatial covariance parameters cannot be estimated consistently. On the other hand, GIC fails to select the true polynomial order consistently under the fixed domain asymptotic framework. Moreover, the growth rate of the domain and the degree of smoothness of candidate regressors in space are shown to play key roles for model selection.

Recommendations

• Consistency of a Class of Information Criteria for Model Selection in Nonlinear Regression
• Dependence of Bayesian model selection criteria and Fisher information matrix on sample size
• Convergence rates of the generalized information criterion
• A consistency property of the AIC for multivariate linear models when the dimension and the sample size are large
• Performance of information criteria for spatial models

Cites work

• scientific article; zbMATH DE number 3483405 (Why is no real title available?)
• scientific article; zbMATH DE number 1219611 (Why is no real title available?)
• scientific article; zbMATH DE number 1034037 (Why is no real title available?)
• scientific article; zbMATH DE number 1423404 (Why is no real title available?)
• Accumulated prediction errors, information criteria and optimal forecasting for autoregressive time series
• Asymptotic optimality for \(C_ p\), \(C_ L\), cross-validation and generalized cross-validation: Discrete index set
• Asymptotic properties of a maximum likelihood estimator with data from a Gaussian process
• Asymptotic properties of criteria for selection of variables in multiple regression
• Asymptotic theory of generalized information criterion for geostatistical regression model selection
• Conditional Akaike information for mixed-effects models
• Consistent covariate selection and post model selection inference in semiparametric regression.
• Consistent procedures for mixed linear model selection
• Estimating the dimension of a model
• Fence methods for mixed model selection
• Fixed-domain asymptotic properties of tapered maximum likelihood estimators
• Interpolation of spatial data. Some theory for kriging
• Maximum likelihood estimation of models for residual covariance in spatial regression
• Model selection in linear mixed models
• On fixed-domain asymptotics and covariance tapering in Gaussian random field models
• Optimal Geostatistical Model Selection
• Order selection for same-realization predictions in autoregressive processes
• Random-Effects Models for Longitudinal Data
• Selecting mixed-effects models based on a generalized information criterion
• Semiparametric Regression
• Semiparametric regression model selections.
• Spatial variation. 2nd ed
• Towards reconciling two asymptotic frameworks in spatial statistics
• Uniform moment bounds of Fisher's information with applications to time series

Cited in

(7)

• Composite likelihood estimation for a Gaussian process under fixed domain asymptotics
• A generalized measure of uncertainty in geostatistical model selection
• Model selection with misspecified spatial covariance structure
• Optimal model average prediction in orthogonal kriging models
• Asymptotic theory of generalized information criterion for geostatistical regression model selection
• Consistency of a Class of Information Criteria for Model Selection in Nonlinear Regression
• Bayesian fixed-domain asymptotics for covariance parameters in a Gaussian process model

Describes a project that uses

Uses Software

• SemiPar

This page was built for publication: Asymptotic theory of generalized information criterion for geostatistical regression model selection

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q482898)