A Bayesian algorithm based on auxiliary variables for estimating GRM with non-ignorable missing data

Abstract

In this paper, a highly effective Bayesian sampling algorithm based on auxiliary variables is used to estimate the graded response model with non-ignorable missing response data. Compared with the traditional marginal likelihood method and other Bayesian algorithms, the advantages of the new algorithm are discussed in detail. Based on the Markov Chain Monte Carlo samples from the posterior distributions, the deviance information criterion and the logarithm of the pseudomarignal likelihood are employed to compare the different missing mechanism models. Two simulation studies are conducted and a detailed analysis of the sexual compulsivity scale data is carried out to further illustrate the proposed methodology.