MLJul 2, 2015
Classical vs. Bayesian methods for linear system identification: point estimators and confidence sets
arXiv:1507.00543v112 citations
AI Analysis
This work addresses system identification for researchers, but it is incremental as it compares existing methods without introducing new ones.
The paper compares classical parametric and Bayesian methods for linear system identification, focusing on point estimators and confidence sets, and reports results from Full Bayes and Empirical Bayes approximations.
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.