A new reference prior for ICAR random effects in Gaussian hierarchical models achieves the same variable selection results as a prior method but reduces computational cost from O(n^3 2^k) to O(n^3) with a proof of equivalence.
Title resolution pending
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
-
A novel reference prior for Gaussian hierarchical models with intrinsic conditional autoregressive random effects
A new reference prior for ICAR random effects in Gaussian hierarchical models achieves the same variable selection results as a prior method but reduces computational cost from O(n^3 2^k) to O(n^3) with a proof of equivalence.