La. Garciacortes et al., USING COUPLING WITH THE GIBBS SAMPLER TO ASSESS CONVERGENCE IN ANIMAL-MODELS, Journal of animal science, 76(2), 1998, pp. 441-447
The coupling method was investigated as a method to assess the converg
ence of the Gibbs sampler when drawing marginal inferences with an ani
mal model. This method is based on the output of two Markov chains wit
h different starting values but the same conditional deviates. The cou
pling method shows that the Gibbs sampler has an exponential convergen
ce when the variance components are assumed to be known. All the varia
bles in the model have the same rate of convergence, and it is closely
related with the largest eigenvalue of a matrix derived from the coef
ficient matrix of the mixed model equations. Models including the vari
ance components as unknowns showed a rate of convergence equal for all
the variables, and this rate of convergence was approximately exponen
tial. The coupling method provides an estimation of the convergence at
the current iteration, and it does not require a post-Gibbs analysis
as do the single chain-based methods. The coupling method is less comp
utationally demanding than multiple chain methods, because only two ch
ains are required to assess convergence.