Bayesian Model with Stan for combining Summary measures
An implementation of Bayesian hierarchical models with Stan for combining the summay measures from multiple sources. In many situations that arise frequently in meta-analysis and small area estimation, among other fields, one is interested in combining estimates from multiple studies or domains given corresponding measures of the uncertainty of each estimate. In such a case, an efficient inference can be accomplished based on a hierarchical Bayesian model with appropriate priors to account for the between-source variation and within source variation. The univariate hierarchical model is used as a general method when there is no obvious correlation between summary estimates and the corresponding uncertainty of the measures. Bivariate model can be thought of as an improvement to the univariate model in terms of estimation accuracy and efficiency in the situation in which measures and their uncertainty are correlated. See "Bivariate Hierarchical Bayesian Model for Combining Summary Measures from Multiple Sources" for more information. The main function of this package is unimod() and bimod().