censlinVarReg {VarReg}

Censored Linear mean and variance regression

Description

censlinVarReg performs censored multivariate mean and multivariate variance regression. This function is designed to be used by the semiVarReg function.

Usage

censlinVarReg(
  dat,
  mean.ind = c(2),
  var.ind = c(2),
  cens.ind = c(3),
  mean.intercept = TRUE,
  para.space = c("all", "positive", "negative"),
  mean.init = NULL,
  var.init = NULL,
  control = list(...),
  ...
)

Arguments

dat

Dataframe containing outcome and covariate data. Outcome data must be in the first column, with censored values set to the limits. Covariates for mean and variance model in next columns.

mean.ind

Vector containing the column numbers of the data in 'dat' to be fit as covariates in the mean model. 0 indicates constant mean option. NULL indicates zero mean option.

var.ind

Vector containing the column numbers of the data in 'dat' to be fit as covariates in the variance model. FALSE indicates constant variance option.

cens.ind

Vector containing the column number of the data in 'dat' to indicate the censored data. 0 indicates no censoring, -1 indicates left (lower) censoring and 1 indicates right (upper) censoring.

mean.intercept

Logical to indicate if an intercept is to be included in the mean model. Default is TRUE.

para.space

Parameter space to search for variance parameter estimates. "positive" means only search positive parameter space, "negative" means search only negative parameter space and "all" means search all. Default is all.

mean.init

Vector of initial estimates to be used for the mean model.

var.init

Vector of initial estimates to be used for the variance model.

control

List of control parameters. See VarReg.control.

...

arguments to be used to form the default control argument if it is not supplied directly

Value

censlinVarReg returns a list of output including:

• converged: Logical argument indicating if convergence occurred.

• iterations: Total iterations performed of the EM algorithm.

• reldiff: the positive convergence tolerance that occured at the final iteration.

• loglik: Numeric variable of the maximised log-likelihood.

• boundary: Logical argument indicating if estimates are on the boundary.

• aic.c: Akaike information criterion corrected for small samples

• aic: Akaike information criterion

• bic: Bayesian information criterion

• hqc: Hannan-Quinn information criterion

• mean.ind: Vector of integer(s) indicating the column number(s) in the dataframe data that were fit in the mean model.

• mean: Vector of the maximum likelihood estimates of the mean parameters.

• var.ind: Vector of integer(s) indicating the column(s) in the dataframe data that were fit in the variance model.

• variance: Vector of the maximum likelihood estimates of the variance parameters.

• cens.ind: Integer indicating the column in the dataframe data that corresponds to the censoring indicator.

• data: Dataframe containing the variables included in the model.