This class implements an Adaptive Multi-site Metropolis random walk algorithm.
R6Class with methods for updating a Node instance.
A multivariate Gaussian proposal is used, for which the proposal variance is a scaled version of the evolving empirical posterior covariance matrix. See Roberts and Rosenthal (2012) Examples of Adaptive MCMC. Journal of Computational and Graphical Statistics. 18:349–367.
Please note that no checks are performed as to the suitability of this algorithm for a particular StochasticNode. It is up to the user to use the correct update algorithm for the appropriate nodes.
the current covariance
the number of updates to burn in
the current tuning matrix
the number of accepted proposals
the number of times
update has been called
the node to which the updater is attached
new(node, tune = rep(0.1, length(node$getData())), burning = 100)
constructor takes an instance of a StochasticNode node, initial tuning vector (diagonal of adaptive tuning matrix), and number of burnin calls.
when called, updates
return the acceptance rate
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.