“…The differences in the model complexity from dif-ferent domains have led to the existence of a wide range of MCMC sampling algorithms. Some of the prominent ones are Metropolis-Hastings algorithm [14]- [16], Gibbs sampling [17]- [19], Langevin MCMC [20]- [22], rejection sampling [23], [24], importance sampling [25], [26], sequential MCMC [27], adaptive MCMC [28], parallel tempering (tempered) MCMC [29]- [32], reversible-jump MCMC [33], [34], specialised MCMC methods for discrete time series models [35]- [37], constrained parameter and model settings [38], [39], and likelihood free MCMC [40]. MCMC sampling methods have also been used for data augmentation [41], [42], model fusion [43], model selection [44], [45], and interpolation [46].…”