Aaron A. King, Ph.D.

Nelson G. Hairston Collegiate Professor of Ecology, Evolutionary Biology,
and Complex Systems, University of Michigan
External Professor, Santa Fe Institute
Fellow of the American Association for the Advancement of Science

Inference for dynamic and latent variable models via iterated, perturbed Bayes maps

E. L. Ionides, D. Nguyen, Y. Atchadé, S. Stoev, and A. A. King
Proceedings of the National Academy of Sciences of the U.S.A. 112(3): 719–724, 2015.

Iterated filtering algorithms are stochastic optimization procedures for latent variable models that recursively combine parameter perturbations with latent variable reconstruction. Previously, theoretical support for these algorithms has been based on the use of conditional moments of perturbed parameters to approximate derivatives of the log likelihood function. Here, a theoretical approach is introduced based on the convergence of an iterated Bayes map. An algorithm supported by this theory displays substantial numerical improvement on the computational challenge of inferring parameters of a partially observed Markov process.

The official version of the paper is here.   Please contact Prof. King if you'd like a reprint.

© 2021 Aaron A. King
3038 Biological Sciences Building
1105 North University Avenue
Ann Arbor MI 48109-1085 USA