2012/06/18 by Marcelo Pereyra, Pereyra, Marcelo, Nicolas Dobigeon +5
Computer Science · Mathematics · #Bayesian Methods and Mixture Models #Computation (stat.CO) #FOS: Computer and information sciences #Markov Chains and Monte Carlo Methods #Methodology (stat.ME) #Statistical Methods and Bayesian Inference #Statistical Methods and Inference
paper · pdf · doi:10.48550/arxiv.1206.3985
openalex publication_date 2012/06/18 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
This report considers the problem of computing the Cramer-Rao bound for the\nparameters of a Markov random field. Computation of the exact bound is not\nfeasible for most fields of interest because their likelihoods are intractable\nand have intractable derivatives. We show here how it is possible to formulate\nthe computation of the bound as a statistical inference problem that can be\nsolve approximately, but with arbitrarily high accuracy, by using a Monte Carlo\nmethod. The proposed methodology is successfully applied on the Ising and the\nPotts models.% where it is used to assess the performance of three state-of-the\nart estimators of the parameter of these Markov random fields.\n