dl2_estimator: DerSimonian and Kacker estimator

In pkimes/upbm: Tools for the analysis of universal protein binding microarrays

Description

This is an implementation of the DerSimonian and Kacker two-step estimator of cross-study variance, originally proposed as an improvement over the one-step DerSimonian and Laird estimator in the context of meta analysis.

Usage

dl2_estimator(Y, vi, k)

Arguments

Y	probe effect sizes
vi	probe variances
k	number of probes

Value

list of estimates:

• betaFE: effect size with no cross-study variance

• varFE: total variance with no cross-study variance

• betaME: effect size with cross-study variance

• varME: total variance with cross-study variance

• tau2: cross-study variance

Author(s)

Patrick Kimes

References

• DerSimonian, R., & Kacker, R. (2007). Random-effects model for meta-analysis of clinical trials: an update. Contemporary Clinical Trials, 28(2), 105-114.

pkimes/upbm documentation built on Oct. 17, 2020, 9:10 a.m.