Parameter expansion to accelerate EM: The PX-EM algorithm

The EM algorithm and its extensions are popular tools for modal estimation but are often criticised for their slow convergence. We propose a new metho d that can often make EM much faster. The intuitive idea is to use a 'covar iance adjustment' to correct the analysis of the M step, capitalising on ex tra information captured in the imputed complete data. The way we accomplis h this is by parameter expansion; we expand the complete-data model while p reserving the observed-data model and use the expanded complete-data model to generate EM. This parameter-expanded EM, PX-EM, algorithm shares the sim plicity and stability of ordinary EM, but has a faster rate of convergence since its M step performs a more efficient analysis. The PX-EM algorithm is illustrated for the multivariate t distribution, a random effects model, f actor analysis, probit regression and a Poisson imaging model.