This class encapsulates the information needed to resample the Monte Carlo chains for the Bayesian Hierarchical Model as described in Myers et al. (2011) using user defined values for \(k\) and \(eta\).
Objects can be created by calls of the form new("mederrResample", log.ir, samp, A, t.new, t.old, grd)
, but most often as the result of a call to bhm.resample
.
log.ir
:Object of class "array"
; logarithm of the importance ratio for each pair of \((k,\eta)\) values.
samp
:Object of class "array"
; resampled MCMC simulation indexes.
A
:Object of class "array"
; transformation ratio for each pair of \((k,\eta)\) values.
t.new
:Object of class "array"
; \(\theta_i\) posterior modes using \((k = \infty, \eta = 1)\).
t.old
:Object of class "numeric"
; \(\theta_i\) posterior modes using user defined \((k, \eta)\) values.
grd
:Object of class "list"
; grid of required \((k, \eta)\) values.
signature(x = "mederrResample", y = "missing")
: : Provides a graphical representation of a mederrResample
object.
Sergio Venturini sergio.venturini@unicatt.it,
Jessica A. Myers jmyers6@partners.org
Myers, J. A., Venturini, S., Dominici, F. and Morlock, L. (2011), "Random Effects Models for Identifying the Most Harmful Medication Errors in a Large, Voluntary Reporting Database". Technical Report.
bayes.rank
.
bhm.constr.resamp
,
bhm.mcmc
.