Take a sneak peek at the new NIST.gov and let us know what you think!
(Please note: some content may not be complete on the beta site.).

View the beta site
NIST logo

Publication Citation: Estimating Heterogeneity Variance in Meta-Analysis

NIST Authors in Bold

Author(s): Andrew L. Rukhin;
Title: Estimating Heterogeneity Variance in Meta-Analysis
Published: August 06, 2012
Abstract: Several new estimators of the between-study variability in a heterogeneous random effects meta-analysis model are derived. One of them can be interpreted as the empirical Bayes procedure for a diffuse prior with the given prior mean. Another is the unbiased estimator which is locally optimal for small values of the parameter. These procedures are compared to the traditional DerSimonian-Laird procedure and the Hedges estimator by means of their mean absolute error, as well as by the quadratic risk of the treatment effect. Confidence intervals are derived by using these estimators and studied via a Monte Carlo study which supports their usage.
Citation: Journal of the Royal Statistical Society Series B-Statistical Methodology
Keywords: Confidence intervals, DerSimonian-Laird estimator, diffuse prior, empirical Bayes approach, heteroscedasticity, random effects model, unbiased estimators.
Research Areas: Statistics