Quantification of Empirical Determinacy: The Impact of Likelihood Weighting on Posterior Location and Spread in Bayesian Meta-Analysis Estimated with JAGS and INLA

The popular Bayesian meta-analysis expressed by normal-normal hierarchical model (NNHM) synthesizes knowledge from several studies and is highly relevant in practice. Moreover, NNHM the simplest (BHM), which illustrates problems typical more complex BHMs. Until now, it has been unclear to what extent data determines marginal posterior distributions of parameters NNHM. To address this issue we computed second derivative Bhattacharyya coefficient with respect weighted likelihood, defined total empirical determinacy (TED), proportion location TED (pEDL), spread (pEDS). We implemented method R package \texttt{ed4bhm} considered two case one simulation study. quantified TED, pEDL pEDS under different modeling conditions such as parametrization, primary outcome, prior. This clarified are determined data. Although these investigations focused on NNHM, proposed applicable generally