confint.hetmeta: Confidence Intervals for 'hetmeta' Objects

In alecri/hetmeta: Heterogeneity Measures in Meta-Analysis

Description

The function calculates confidence intervals for the heterogeneity measures in a 'hetmeta' object.

Usage

## S3 method for class 'hetmeta'
confint(object, parm, level, rma.type = FALSE, ...)

Arguments

object	an object of class hetmeta produced by hetmeta.
parm	this argument is here for compatability with the generic function confint, but is (currently) ignored.
level	numerical value between 0 and 100 specifying the confidence interval level (if unspecified, the default is to take the value from the object).
rma.type	to obtain the confidence interval for I2 as in 'confint.rma.uni'.
...	further arguments passed to or from other methods.

Details

The confidence intervals are constructed based on the (asymptotic) normal distribution of the estimators. Standard error are derived using the delta method. See the references for more details.

Value

The confint function returns a matrix with the estimate, standard error, and confidence intervals for the heterogeneity measures estimated by "hetmeta" (R_b, I^2, R_I, and CV_b)

Author(s)

Alessio Crippa, [e-mail]

References

Takkouche B, Khudyakov P, Costa-Bouzas J, Spiegelman D. Confidence Intervals for Heterogeneity Measures in Meta-analysis. Am. J. Epidemiol. 2013:kwt060.

Crippa A, Khudyakov P, Wang M, Orsini N, Spiegelman D. A new measure of between-studies heterogeneity in meta-analysis. 2016. Stat. Med. In Press.

See Also

hetmeta

Examples

## load BCG vaccine data
data(dat.bcg)

## random-effects model of log relative risks
dat <- escalc(measure="RR", ai=tpos, bi=tneg, ci=cpos, di=cneg, data=dat.bcg)
res <- rma(yi, vi, data=dat)

## heterogeneity measures
het <- hetmeta(res)
confint(het)

alecri/hetmeta documentation built on July 31, 2019, 7:37 a.m.