multiplier

Multiplier to scale prior variance

Specification

  • Alias: None

  • Arguments: None

  • JSON argument key: None

  • Default: 1.0

Description

The initial proposal covariance will be given by the prior variance times the multiplier.

Default Behavior When using prior-based proposal covariance, the default is to use the prior variance for the proposal (multiplier = 1.0)

Usage Tips

The prior variance may result in too tight or narrow a proposal covariance. The multiplier can be used to scale all entries of the prior variance in determining the initial proposal covariance.

Examples

method
  bayes_calibration queso
    samples = 2000 seed = 348
    dram
    proposal_covariance prior
      multiplier 0.1
{
  "method": [
    {
      "bayes_calibration": {
        "queso": {
          "samples": 2000,
          "seed": 348,
          "dram": {},
          "proposal_covariance": {
            "prior": {"multiplier": 0.1}
          }
        }
      }
    }
  ]
}