A comparison of estimation methods for multilevel models of spatially structured data

被引:24
|
作者
Bivand, Roger [1 ]
Sha, Zhe [2 ]
Osland, Liv [3 ]
Thorsen, Ingrid Sandvig [4 ,5 ]
机构
[1] Norwegian Sch Econ, Helleveien 30, N-5045 Bergen, Norway
[2] Univ Oxford, 24-29 St Giles, Oxford OX1 3LB, England
[3] Stord Haugesund Univ Coll, Bjornsonsgt 45, N-5528 Haugesund, Norway
[4] Univ Bergen, Dept Math, Thormohlensgate 41A, N-5006 Bergen, Norway
[5] Analysetjenester AS, Thormohlensgate 41A, N-5006 Bergen, Norway
关键词
Spatially structured random effects; Spatial multilevel models; Estimation of CAR and SAR models; CARLO MAXIMUM-LIKELIHOOD; BAYESIAN-INFERENCE; CONVERGENCE; DEPENDENCE; PACKAGE;
D O I
10.1016/j.spasta.2017.01.002
中图分类号
P [天文学、地球科学];
学科分类号
07 ;
摘要
Two recent contributions (Dong et al., 2015; Osland et al., 2016) point to the relevance of multilevel models for spatially structured data. In Osland et al. (2016) these models are used to examine the importance of district-level covariates for house prices in Stavanger, Norway, in Dong et al. (2015) similarly for land parcel prices in Beijing; we use these data sets in our comparison. In Osland et al. (2016), a district-level spatial random effect was fitted using an intrinsic CAR model estimated using WinBUGS. Dong et al. (2015) used R code provided in supplementary materials to their article, and subsequently improved in an R package (Dong et al., 2016a); computation there used custom MCMC C++ code to fit a SAR district-level spatial random effect. This article compares approaches to estimating models of this kind, using the R packages R2WinBUGS, HSAR, INLA, R2BayesX, hglm and the new package mclcar for Monte Carlo maximum likelihood estimation (Sha, 2016b). We show that multilevel models of spatially structured data may be estimated readily using a variety of approaches, not only the intrinsic CAR model more typically found in the existing literature. We also point to a range of issues for further research in situations in which data acquired at different levels of spatial resolution are combined. (C) 2017 Elsevier B.V. All rights reserved.
引用
收藏
页码:440 / 459
页数:20
相关论文
共 50 条
  • [1] A Comparison of Estimation Methods for Multilevel Logistic Models
    Mirjam Moerbeek
    Gerard J. P. Van Breukelen
    Martijn P. F. Berger
    [J]. Computational Statistics, 2003, 18 : 19 - 37
  • [2] A comparison of estimation methods for multilevel logistic models
    Moerbeek, M
    Van Breukelen, GJP
    Berger, MPF
    [J]. COMPUTATIONAL STATISTICS, 2003, 18 (01) : 19 - 37
  • [3] Fast methods for spatially correlated multilevel functional data
    Staicu, Ana-Maria
    Crainiceanu, Ciprian M.
    Carroll, Raymond J.
    [J]. BIOSTATISTICS, 2010, 11 (02) : 177 - 194
  • [4] Are multiple regression models of spatially structured data to be trusted?
    Hawkins, Bradford A.
    [J]. JOURNAL OF BIOGEOGRAPHY, 2012, 39 (05) : 998 - 998
  • [5] A comparison of errors in variables methods for use in regression models with spatially misaligned data
    Lopiano, Kenneth K.
    Young, Linda J.
    Gotway, Carol A.
    [J]. STATISTICAL METHODS IN MEDICAL RESEARCH, 2011, 20 (01) : 29 - 47
  • [6] Applications of Multilevel Structured Additive Regression Models to Insurance Data
    Lang, Stefan
    Umlauf, Nikolaus
    [J]. COMPSTAT'2010: 19TH INTERNATIONAL CONFERENCE ON COMPUTATIONAL STATISTICS, 2010, : 155 - 164
  • [7] Multilevel Conditional Autoregressive models for longitudinal and spatially referenced epidemiological data
    Djeudeu, D.
    Moebus, S.
    Ickstadt, K.
    [J]. SPATIAL AND SPATIO-TEMPORAL EPIDEMIOLOGY, 2022, 41
  • [8] Multilevel modeling in the presence of outliers: A comparison of robust estimation methods
    Finch, Holmes
    [J]. PSICOLOGICA, 2017, 38 (01): : 57 - 92
  • [9] Spatially structured discrete population models
    Hiebeler, DE
    [J]. ADVANCED TOPICS IN BIOMATHEMATICS, 1998, : 79 - 83
  • [10] Scaling structured interview data: A comparison of two methods
    Cerel, J
    Fristad, MA
    [J]. JOURNAL OF THE AMERICAN ACADEMY OF CHILD AND ADOLESCENT PSYCHIATRY, 2001, 40 (03): : 341 - 346