dirichlet.iteration: dirichlet.iteration

In kreidles/informativeDropout: informativeDropout: varying coefficient models to account for informative dropout

Description

dirichlet.iteration

Usage

dirichlet.iteration(weights.mixing = NULL, weights.conditional = NULL,
  cluster.assignments = NULL, betas = NULL, betas.deviations = NULL,
  betas.covariates = NULL, betas.covariates.mu = NULL,
  betas.covariates.sigma = NULL, dp.dist.mu0 = NULL,
  dp.dist.sigma0 = NULL, dp.cluster.sigma = NULL,
  dp.concentration = NULL, cluster.N = NULL, cluster.mu = NULL,
  sigma.error = NULL, expected.intercept = NULL,
  expected.slope = NULL, intercept.dropoutTimes = NULL,
  slope.dropoutTimes = NULL, intercept.dropoutTimes.censored = NULL,
  slope.dropoutTimes.censored = NULL, density.intercept = NULL,
  density.slope = NULL)

Arguments

weights.mixing	vector containing the probability of belonging to cluster k, for k = 1 to the number of clusters
weights.conditional	vector containing the probability of belonging to cluster k, given that the subject was not in clusters 1 to k - 1
cluster.assignments	current cluster assignments
betas	A (k x 3) matrix of regression coefficients for the random intercept, slope, and log of the dropout time for each cluster, with k = number of clusters
betas.deviations	An (N x k) matrix of subject specific deviations from the cluster means
betas.covariates	A (c x 1) vector of regression coefficients for covariates, with c = number of covariates
betas.covariates.mu	A (c x 1) vector representing the mean of the distribution of regression coefficients related to covariates
betas.covariates.sigma	A (c x c) vector representing the covariance of the distribution of regression coefficients related to covariates
dp.dist.mu0	A (3 x 1) vector of means for the baseline distribution of the Dirichlet process
dp.dist.sigma0	A (3 x 3) matrix representing the covariance of the baseline distribution of the Dirichlet process
dp.cluster.sigma	A (3 x 3) matrix representing the covariance of the random intercept, slope, and log dropout time for each cluster. This covariance is the same for each cluster
dp.concentration	the concentration parameter (alpha) for the Dirichlet process
cluster.N	A (k x 1) vector indicating the number of subjects in cluster k, for k = 1 to the number of clusters
cluster.mu	A (k x 3) matrix of means for the random intercept, slope, and log dropout time in cluster k, for k = 1 to the number of clusters
sigma.error	For Gaussian outcomes, the residual error variance
expected.intercept	the expected value of the random intercept
expected.slope	the expected value of the random slope
intercept.dropoutTimes	estimated intercepts by dropout times
slope.dropoutTimes	estimated slopes by dropout times
intercept.dropoutTimes.censored	estimated intercepts by censored dropout times
slope.dropoutTimes.censored	estimated slopes by censored dropout times
density.intercept	estimated density of the random intercepts
density.slope	estimated density of the random slopes
expected.intercept	expected value of the random intercepts
expected.slope	expected value of the random slopes

kreidles/informativeDropout documentation built on Sept. 13, 2020, 12:15 a.m.