High-dimensional Inference and FDR Control for Simulated Markov Random Fields

2022/02/11 by Wei, Haoyu, Lei, Xiaoyu, Han, Yixin +1
#FOS: Computer and information sciences #FOS: Mathematics #Machine Learning (cs.LG) #Machine Learning (stat.ML) #Statistics Theory (math.ST)

paper · doi:10.48550/arxiv.2202.05612

Abstract

Identifying important features linked to a response variable is a fundamental task in various scientific domains. This article explores statistical inference for simulated Markov random fields in high-dimensional settings. We introduce a methodology based on Markov Chain Monte Carlo Maximum Likelihood Estimation (MCMC-MLE) with Elastic-net regularization. Under mild conditions on the MCMC method, our penalized MCMC-MLE method achieves ℓ₁-consistency. We propose a decorrelated score test, establishing both its asymptotic normality and that of a one-step estimator, along with the associated confidence interval. Furthermore, we construct two false discovery rate control procedures via the asymptotic behaviors for both p-values and e-values. Comprehensive numerical simulations confirm the theoretical validity of the proposed methods.

Related

• Multilevel Markov Chain Monte Carlo with likelihood scaling for Bayesian inversion with high-resolution observations
• Reduced-dimensional Monte Carlo Maximum Likelihood for Latent Gaussian Random Field Models
• Structure Learning of Markov Random Fields through Grow-Shrink Maximum Pseudolikelihood Estimation
• Bayesian Structure Learning for Markov Random Fields with a Spike and Slab Prior
• Bayesian Structure Learning for Markov Random Fields with a Spike and Slab Prior