Work

Robust Model Order Selection via Dithered Differential Step-Down Thresholding

Year: 2026

Type: preprint

Abstract: We propose a differential step-down thresholding algorithm with randomized grid dithering for robust model order selection. It detects the value clustering of noise extrema. Our algorithm asymptotically achieves exact order recovery and prevents false alarms even under lowered detection thresholds.

Source: arXiv (Cornell University)

DOI: 10.48550/arxiv.2608.29868

Author: Kharin, Aleksandr

Institutions: National Research Tomsk State University, Laboratoire de Mathématiques Raphaël Salem, Université de Rouen Normandie

Language: English

Topic: Statistical Methods and Inference

Subfield: Statistics and Probability

Field: Mathematics

Domain: Physical Sciences

Open Access status: green

Funder: Russian Science Foundation

Award: G3356102944

OpenAlex ID: W7204942354

Cites:

Cited by:

Related to: 10