A1182
Title: Maximum contribution to the likelihood: An estimation approach for stochastic expectation-maximization algorithm
Authors: Alexander Sharp - University of Waterloo (Canada) [presenting]
Ryan Browne - University of Waterloo (Canada)
Abstract: The stochastic EM algorithm replaces the E-step with a Monte Carlo approximation, trading monotonicity for the potential to escape local maxima. Estimation techniques include averaging the tail of the chain and choosing the value in the chain associated with the largest likelihood value. It is demonstrated that the latter estimator diverges from the maximum likelihood estimate with high probability as the dimensionality of the parameter increases but that it is also more precise in terms of chain length when the parameter is a scalar. Based on these findings, a new estimator is proposed which achieves this same level of precision for the inference of multidimensional parameters is proved. Simulation studies demonstrate the benefits of the proposed estimator when compared to topical approaches.
