shape_adapter: Create object to adapt proposal shape.

In rmcmc: Robust Markov Chain Monte Carlo Methods

Create object to adapt proposal shape.

Description

Create object to adapt proposal shape.

Usage

shape_adapter(type = "covariance", kappa = 1)

Arguments

type	Type of shape adapter to use. One of: "variance": Diagonal shape matrix adaptation based on estimates of target distribution variances (see variance_shape_adapter()), "covariance": Dense shape matrix adaptation based on estimates of target distribution covariance matrix (see covariance_shape_adapter()).
kappa	Decay rate exponent in ⁠[0.5, 1]⁠ for adaptation learning rate. Value of 1 (default) corresponds to computing empirical (co)variances.

Value

List of functions with entries

• initialize, a function for initializing adapter state and proposal parameters at beginning of chain,

• update a function for updating adapter state and proposal parameters on each chain iteration,

• finalize a function for performing any final updates to adapter state and proposal parameters on completion of chain sampling (may be NULL if unused).

• state a zero-argument function for accessing current values of adapter state variables.

See Also

variance_shape_adapter(), covariance_shape_adapter()

Examples

proposal <- barker_proposal()
adapter <- shape_adapter()
adapter$initialize(proposal, chain_state(c(0, 0)))

rmcmc documentation built on April 3, 2025, 5:27 p.m.