Set the MCMC sampler.
The sampler determines the type of jumping rule to be used when running the MCMC analysis.
sampler (str or sherpa.sim.Sampler instance) – When a string, the name of the sampler to use (case insensitive). The supported options are given by the list_samplers function.
The jumping rules are:
The Metropolis-Hastings rule, which jumps from the best-fit location, even if the previous iteration had moved away from it.
This is the Metropolis with Metropolis-Hastings algorithm, that jumps from the best-fit with probability
p_M, otherwise it jumps from the last accepted jump. The value of
p_Mcan be changed using set_sampler_opt.
This is used when the effective area calibration uncertainty is to be included in the calculation. At each nominal MCMC iteration, a new calibration product is generated, and a series of N (the
nsubitersoption) MCMC sub-iteration steps are carried out, choosing between Metropolis and Metropolis-Hastings types of samplers with probability
p_M. Only the last of these sub-iterations are kept in the chain. The
p_Mvalues can be changed using set_sampler_opt.
Another sampler for use when including uncertainties due to the effective area.