mederrFit-class {mederrRank}

Class "mederrFit". Simulated Monte Carlo Chains (Step 1) for the Bayesian Hierarchical Model Used to Identify the Most Harmful Medication Errors.

Description

This class encapsulates the simulated Monte Carlo chains for the Bayesian Hierarchical Model as described in Myers et al. (2011) forcing a symmetric normal distribution on the \theta_i, i=1,\ldots,n.

Objects from the Class

Objects can be created by calls of the form new("mederrFit", thetai, deltaj, gamma, sigma2, tau2, p.acc.i, p.acc.j, tune.theta, tune.delta, k, eta), but most often as the result of a call to bhm.mcmc or to bhm.constr.resamp.

Slots

thetai:

Object of class "matrix"; simulated chains for the \theta_i, i = 1,\ldots,n, error profiles random effects; see bhm.mcmc.

deltaj:

Object of class "matrix"; simulated chains for the \delta_j, i = j,\ldots,J, hospitals random effects; see bhm.mcmc.

gamma:

Object of class "numeric"; simulated chain for the \gamma parameter; see bhm.mcmc.

sigma2:

Object of class "numeric"; simulated chain for the \sigma^2 parameter; see bhm.mcmc.

tau2:

Object of class "numeric"; simulated chain for the \tau^2 parameter; see bhm.mcmc.

p.acc.i:

Object of class "numeric"; acceptance rates for the error profiles random effects.

p.acc.j:

Object of class "numeric"; acceptance rates for the hospitals random effects.

tune.theta:

Object of class "numeric"; last updated values of the \theta_i working variances for the Metropolis step.

tune.delta:

Object of class "numeric"; last updated values of the \delta_j working variances for the Metropolis step.

k:

Object of class "numeric"; k value used in the simulation.

eta:

Object of class "numeric"; \eta value used in the simulation.

Methods

plot

signature(x = "mederrFit", y = "mederrFit"): Provides a graphical representation of the estimates.

summary

signature(object = "mederrFit"): Summarizes the information regarding the estimates.

Author(s)

Sergio Venturini [e-mail],

Jessica A. Myers [e-mail]

References

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.

See Also

bayes.rank, bhm.constr.resamp, bhm.mcmc.