calib_ests: Return calibrated estimates of studies' population effect...

In MetaUtility: Utility Functions for Conducting and Interpreting Meta-Analyses

Description

Returns estimates of the population effect in each study based on the methods of Wang & Lee (2019). Unlike the point estimates themselves, these "calibrated" estimates have been appropriately shrunk to correct the overdispersion that arises due to the studies' finite sample sizes. By default, this function uses Dersimonian-Laird moments-based estimates of the mean and variance of the true effects, as Wang & Lee (2019) recommended.

Usage

calib_ests(yi, sei, method = "DL")

Arguments

yi	Vector of study-level point estimates
sei	Vector of study-level standard errors
method	Estimation method for mean and variance of population effects (passed to metafor::rma.uni)

References

Wang C-C & Lee W-C (2019). A simple method to estimate prediction intervals and predictive distributions: Summarizing meta-analyses beyond means and confidence intervals. Research Synthesis Methods.

Examples

d = metafor::escalc(measure="RR", ai=tpos, bi=tneg,
                     ci=cpos, di=cneg, data=metadat::dat.bcg)

# calculate calibrated estimates
d$calib = calib_ests( yi = d$yi,
                      sei = sqrt(d$vi) )

# look at 5 studies with the largest calibrated estimates
d = d[ order(d$calib, decreasing = TRUE), ]
d$trial[1:5]

# look at kernel density estimate of calibrated estimates
plot(density(d$calib))

MetaUtility documentation built on Oct. 30, 2021, 5:07 p.m.