The limiting distribution of the Gibbs sampler for the intrinsic conditional autoregressive model (Q2330484)

After quite an extensive presentation of the problem, the second and third sections of the article are dedicated to the description of the intrinsic conditional autoregressive (ICAR) model and to the one-at-a-time Gibbs sampler centering on the fly (OGC), used to simulate the vector of random effects from the ICAR model. The sampler OGC-algorithm is presented. The author proves that the limiting distribution of the OGC algorithm is the same as the distribution proposed in the paper of \textit{M. J. Keefe} et al. [``On the formal specification of sum-zero constrained intrinsic conditional autoregressive models'', Spat. Stat. 24, 54--65 (2018; \url{doi:10.1016/j.spasta.2018.03.007})].