Period at which to update derivative-based proposal covariance

Specification

Alias: none

Argument(s): INTEGER

Description

For derivative-based proposal covariance, this specifies the period (number of accepted MCMC samples) after which the proposal covariance is updated using derivative values at the current chain point.

Default Behavior

When update_period is not specified, derivatives will inform the proposal covariance at the start of the chain, but not updated further.

Usage Tips

The update_period should be tailored to the size of the total chain, accounting for the relative expense of derivative-based proposal updates.

Examples

method,
        bayes_calibration queso
          samples = 2000 seed = 348
          delayed_rejection
          emulator pce sparse_grid_level = 2
          proposal_covariance derivatives
            update_period = 40  # update proposal covariance every 40 points