A0445
Title: Graphical and numerical diagnostic tools to assess multiple imputation models by posterior predictive checking
Authors: Mingyang Cai - Utrecht University (China) [presenting]
Abstract: A method is proposed to diagnose imputation models based on posterior predictive checking. To assess the congeniality of imputation models, we compare the observed data with their replicates generated under corresponding posterior predictive distributions. The idea is that if the imputation model is congenial with the substantive model, the observed data is expected to locate in the centre of corresponding predictive posterior distributions. We investigate the proposed diagnostic method for parametric and non-parametric imputation approaches, continuous and discrete incomplete variables, univariate and multivariate missingness patterns. The results show the validity of the proposed diagnostic method.
