metagen (version 1.0)

hConfidence: Inference: Based on methods of moments and maximum likelihood.

Description

Calculates the so called Q-profiling confidence interval for the heterogeneity for data following a random effects meta regression model.

Usage

hConfidence(y, d, x, sgnf)

Arguments

y

k-vector of study responses.

d

k-vector of heteroscedasticity.

x

design k-p-matrix.

sgnf

significance levels.

Value

A data frame containing the bounds of the interval estimate.

Examples

bcg   <- bcgVaccineData()
bcg_y <- bcg$logrisk
bcg_d <- bcg$sdiv
bcg_s <- bcg$size
bcg_x <- cbind(1,bcg$x)
sgnf_lev <- c(0.01, 0.025, 0.05, 0.01)

set.seed(865287113) # for reproducibility

hConfidence(y=bcg_y, d=bcg_d, x=bcg_x, sgnf=0.025)
hConfidence(y=bcg_y, d=bcg_d, x=bcg_x, sgnf=sgnf_lev)