glmLimits: glmLimits function

Description Usage Arguments Value Examples

Description

This function allows you to adjust multiple Bayesian GLM models with predictors subjected to a limit of detection (LOD) specified with a BUGS-language description of the prior distribution, and a set of data.

Usage

1
2
3
4
5
glmLimits(formula, data, LimitVariables, lod = NULL, loq = NULL,
  family = "gaussian", logLimit = FALSE, n.iter = 1000, n.chains = 1,
  alpha.corrected = 0.05, quantiles = unique(c(alpha.corrected/2, 0.025,
  0.975, 1 - (alpha.corrected/2))), quiet = FALSE, progress.bar = "text",
  dic.type = "popt", n.adapt = 1000, ...)

Arguments

formula

an object of class "formula": a symbolic description of the model to be fitted.

data

a data.frame containing the variables in the model.

LimitVariables

character string of names of the variables subjected to a limit.

lod

numeric vector with the same length of LimitVariables containing the lower limit values for the variables with LOD (with the same order).

loq

optional numeric vector with the same length of LimitVariables containing the upper limit values for the variables with limits (with the same order)

family

character with a description of the error distribution to be used in the model. Possible values are "gaussian" (default) for continuous outcomes or "binomial" for binary outcomes.

logLimit

a logical indicating if limit variables would be assumed distributed as log-normal.

n.iter

number of iterations of the Markov chain to run to be used in rjags::update(), rjags::coda.samples() and rjags::dic.samples().

n.chains

the number of parallel chains for the model to be used in rjags::jags.model().

alpha.corrected

significance level to be used for the confident intervals.

quantiles

numeric vector containing the percentiles in [0,1] to obtain of the estimated parameters.

progress.bar

type of progress bar. Possible values are "text" (default), "gui", and "none" to be used in rjags::update().

dic.type

type of penalty to use in rjags::dic.samples().

n.adapt

the number of iterations for adaptation to be used in rjags::jags.model(). If n.adapt = 0 then no adaptation takes place.

Value

It returns an object of class "glmLimits" containing a model of class "jags".

Examples

1
glmLimits(formula, data, LimitVariables, lod = NULL, loq = NULL, family = "gaussian", logLimit = FALSE, n.iter = 1000, n.chains = 1, alpha.corrected = 0.05, quantiles = unique(c(alpha.corrected/2, 0.025, 0.975, 1-(alpha.corrected/2))), quiet = FALSE, progress.bar = "text", dic.type = "popt", n.adapt = 1000, ...)

isglobal-brge/lodregression documentation built on May 18, 2019, 5:50 a.m.