dram
Use the DRAM MCMC algorithm
Topics
bayesian_calibration
Specification
Alias: None
Arguments: None
JSON argument key: None
Child Keywords:
Required/Optional |
Description of Group |
JSON Group Key |
Dakota Keyword |
Dakota Keyword Description |
|---|---|---|---|---|
Optional |
Number of stages |
|||
Optional |
Type of scaling to use |
|||
Optional |
Scaling parameter |
|||
Optional |
Number of steps between updates of the proposal covariance |
|||
Optional |
Number of steps prior to start of proposal covariance adaptation |
|||
Optional |
Sample covariance scaling used to define proposal covariance |
Description
The type of Markov Chain Monte Carlo used. This keyword specifies the use of DRAM, (Delayed Rejection Adaptive Metropolis) [HLMS06].
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 knows very little about the proposal covariance, using dram is a recommended strategy. The proposal covariance is adaptively updated, and the delayed rejection may help improve low acceptance rates.
Examples
method,
bayes_calibration queso
dram
samples = 10000 seed = 348
{
"method": [
{
"bayes_calibration": {
"queso": {
"dram": {},
"samples": 10000,
"seed": 348
}
}
}
]
}

