variance_shape_adapter {rmcmc}R Documentation

Create object to adapt proposal with per dimension scales based on estimates of target distribution variances.

Description

Corresponds to variance variant of Algorithm 2 in Andrieu and Thoms (2009), which is itself a restatement of method proposed in Haario et al. (2001).

Usage

variance_shape_adapter(kappa = 1)

Arguments

kappa

Decay rate exponent in ⁠[0.5, 1]⁠ for adaptation learning rate. Value of 1 (default) corresponds to computing empirical variances.

Value

List of functions with entries

References

Andrieu, C., & Thoms, J. (2008). A tutorial on adaptive MCMC. Statistics and Computing, 18, 343-373.

Haario, H., Saksman, E., & Tamminen, J. (2001). An adaptive Metropolis algorithm. Bernoulli, 7(2): 223-242.

Examples

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

[Package rmcmc version 0.1.1 Index]