Dakota Reference Manual  Version 6.4
Large-Scale Engineering Optimization and Uncertainty Analysis
 All Pages
metropolis_hastings


Use the Metropolis-Hastings MCMC algorithm

Topics

This keyword is related to the topics:

Specification

Alias: none

Argument(s): none

Default: dram

Description

This keyword specifies the use of a Metropolis-Hastings algorithm for the MCMC chain generation. This means there is no delayed rejection and no adaptive proposal covariance updating as in DRAM.

Default Behavior

Five MCMC algorithm variants are supported: dram, delayed_rejection, adaptive_metropolis, metropolis_hastings, and multilevel. The default is dram.

Usage Tips

If the user wants to use Metropolis-Hastings, possibly as a comparison to the other methods which involve more chain adaptation, this is the MCMC type to use.

Examples

method,
        bayes_calibration queso
          metropolis_hastings 
          samples = 10000 seed = 348