Derives closed-form m* threshold below which spatial random effects materially affect regression inference in multilevel areal data and above which nonspatial models suffice, with O(m^{-1}) convergence in posterior variance differences.
IEEE Transactions on Pattern Analysis and Machine Intelligence , volume=
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
stat.ME 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
On the Need for Spatial Random Effects in Bayesian Regression Models for Multilevel Areal Data
Derives closed-form m* threshold below which spatial random effects materially affect regression inference in multilevel areal data and above which nonspatial models suffice, with O(m^{-1}) convergence in posterior variance differences.