Sign-Perturbed Sums: A New System Identification Approach for Constructing Exact Non-Asymptotic Confidence Regions in Linear Regression Models

DOI10.1109/TSP.2014.2369000zbMath1394.94141arXiv1807.08216MaRDI QIDQ4579668

Balázs Csanád Csáji, Marco C. Campi, Erik Weyer

Publication date: 22 August 2018

Published in: IEEE Transactions on Signal Processing (Search for Journal in Brave)

Full work available at URL: https://arxiv.org/abs/1807.08216

Mathematics Subject Classification ID

Linear regression; mixed models (62J05) Signal theory (characterization, reconstruction, filtering, etc.) (94A12)

Related Items (9)

Asymptotic properties of SPS confidence regions ⋮ Non-asymptotic state-space identification of closed-loop stochastic linear systems using instrumental variables ⋮ On the analysis of two theoretical approaches to predict the material strength under dynamic loading ⋮ Facing undermodelling in sign-perturbed-sums system identification ⋮ On the possibility of using the method of sign-perturbed sums for the processing of dynamic test data ⋮ Non-asymptotic confidence regions for the parameters of EIV systems ⋮ Distribution-free uncertainty quantification for kernel methods by gradient perturbations ⋮ Bayesian frequentist bounds for machine learning and system identification ⋮ A simple condition for the boundedness of sign-perturbed-sums (SPS) confidence regions

This page was built for publication: Sign-Perturbed Sums: A New System Identification Approach for Constructing Exact Non-Asymptotic Confidence Regions in Linear Regression Models