EUVIMEDEUVIMED entdecken
Crossref

High-dimensional inference and FDR control for simulated Markov random fields

Haoyu Wei, Xiaoyu Lei, Yixin Han, Huiming Zhang

Statistics and Its Interface · 2026 · 19 (3) · 369-391

DOI: 10.4310/sii.260602005242 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Statistics and Its Interface, 2026. Verfasst von Haoyu Wei und 3 weiteren Autor:innen.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

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 $\ell_{1}$-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.

Zitieren

APA

Wei, H., Lei, X., Han, Y., & Zhang, H. (2026). High-dimensional inference and FDR control for simulated Markov random fields. Statistics and Its Interface, 19(3), 369-391. https://doi.org/10.4310/sii.260602005242

Vancouver

Wei H, Lei X, Han Y, Zhang H. High-dimensional inference and FDR control for simulated Markov random fields. Statistics and Its Interface. 2026;19(3):369-391. doi:10.4310/sii.260602005242

BibTeX
@article{Wei2026,
  author = {Wei, Haoyu and Lei, Xiaoyu and Han, Yixin and Zhang, Huiming},
  title = {High-dimensional inference and FDR control for simulated Markov random fields},
  journal = {Statistics and Its Interface},
  year = {2026},
  volume = {19},
  number = {3},
  pages = {369--391},
  doi = {10.4310/sii.260602005242}
}
Kostenlos starten →