Premium
Quo vadis, Bayesian identification?
Author(s) -
Kulhavý Rudolf,
Ivanova Petya
Publication year - 1999
Publication title -
international journal of adaptive control and signal processing
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.73
H-Index - 66
eISSN - 1099-1115
pISSN - 0890-6327
DOI - 10.1002/(sici)1099-1115(199908)13:6<469::aid-acs555>3.0.co;2-o
Subject(s) - bayesian probability , computer science , identification (biology) , smoothing , algorithm , parametric statistics , gaussian , monte carlo method , mathematical optimization , mathematics , artificial intelligence , data mining , econometrics , statistics , botany , biology , physics , quantum mechanics
The Bayesian identification of non‐linear, non‐Gaussian, non‐stationary or non‐parametric models is notoriously known as computer‐intensive and not solvable in a closed form. The paper outlines three major approaches to approximate Bayesian estimation, based on locally weighted smoothing of data, iterative and non‐iterative Monte Carlo simulation and direct approximation of an information ‘distance’ between the empirical and model distributions of data. The information‐based view of estimation is used throughout to give more insight into the methods and show their mutual relationship. Copyright © 1999 John Wiley & Sons, Ltd.