From: A survey of Monte Carlo methods for parameter estimation
P3 Adaptation by using MCMC transitions to obtain \(\{\boldsymbol {\mu }_{n,t}\}_{n=1}^{N}\) given \(\{\boldsymbol {\mu }_{n,t-1}\}_{n=1}^{N}\), as in [273, 283–286]. |
P4 Adaptation by using stochastic gradient search of π for moving \(\{\boldsymbol {\mu }_{n,t-1}\}_{n=1}^{N}\) to \(\{\boldsymbol {\mu }_{n,t}\}_{n=1}^{N}\) [278, 281]. |