Figure 2
From: Goodness-of-fit testing for meta-analysis of rare binary events
Comparison of bias, MSE and coverage results for Bayesian estimates of \(\theta _{0}\), \(\tau ^{2}\) and \(\rho\) using the Inverse-Wishart prior (IW), the Hierarchical Inverse-Wishart prior (HIW), the Scaled Inverse-Wishart prior (SIW) and the independent prior (IND) for meta-analysis of rare binary events with \(I=20\) studies. For each setting, results were generated using 200 replicates and the nominal coverage level was set as \(95\%\).
