Classical vs. Bayesian methods for linear system identification: point estimators and confidence sets
Machine Learning
2015-07-03 v1
Abstract
This paper compares classical parametric methods with recently developed Bayesian methods for system identification. A Full Bayes solution is considered together with one of the standard approximations based on the Empirical Bayes paradigm. Results regarding point estimators for the impulse response as well as for confidence regions are reported.
Cite
@article{arxiv.1507.00543,
title = {Classical vs. Bayesian methods for linear system identification: point estimators and confidence sets},
author = {D. Romeres and G. Prando and G. Pillonetto and A. Chiuso},
journal= {arXiv preprint arXiv:1507.00543},
year = {2015}
}
Comments
number of pages = 8, number of figures = 4