Simulations for the Q statistic with constant and inverse variance weights for binary effect measures

06/17/2022
by   Elena Kulinskaya, et al.
0

Cochran's Q statistic is routinely used for testing heterogeneity in meta-analysis. Its expected value (under an incorrect null distribution) is part of several popular estimators of the between-study variance, τ^2. Those applications generally do not account for estimation of the variances in the inverse-variance weights. Importantly, those weights make approximating the distribution of Q (more explicitly, Q_IV) rather complicated. As an alternative, we are investigating a Q statistic, Q_F, whose constant weights use only the studies' effective sample sizes. For log-odds-ratio, log-relative-risk, and risk difference as the measure of effect, we study, by simulation, approximations to distributions of Q_IV and Q_F, as the basis for tests of heterogeneity. Results of our simulations are provided and in 114 A4 Figures, 133 pages in total.

READ FULL TEXT

Please sign up or login with your details

Forgot password? Click here to reset