Connecting informative experiments, the information matrix and the minima of a Prediction Error Identification criterion
(2009) 15th IFAC Symposium on System Identification, SYSID 2009 — Location: Saint-Malo (6.July.2009)
Files
No attached file found for this publication.
Details
- Authors
- Abstract
- This paper establishes, in a Prediction Error Identification (PEI) context, the connections that exist between the identifiability of the model structure, the informativity of the data, the information matrix and the existence of a unique global minimum of the PEI criterion. By introducing the concept of informative data at a particular parameter value, we are able to establish a number of equivalences and connections between these four ingredients of the identification problem, for both open-loop and closed-loop identification. © 2009 IFAC.
- Affiliations
Citations
Gevers, M., Bazanella, A. S., & Bombois, X. (2009). Connecting informative experiments, the information matrix and the minima of a Prediction Error Identification criterion. 15th IFAC Symposium on System Identification, SYSID 2009, 675-680.