0312688 - ÚTIA 2009 RIV SI eng C - Conference Paper (international conference)
Dedecius, Kamil
Partial Forgetting in Autoregression Models.
[Parciální zapomínání v autoregresních modelech.]
Proceedings of the 9th International PhD Workshop on Systems and Control, Young Generation Viewpoint. Izola: Jozef Stefan Institute, 2008 - (Gašperin, M.; Pregelj, B.), s. 1-6. ISBN 978-961-264-003-3.
[9th International PhD Workshop on Systems and Control: A Young Generation Viewpoint. Izola (SI), 01.10.2008-03.10.2008]
Institutional research plan: CEZ:AV0Z10750506
Keywords : autoregression model * forgetting * partial forgetting * estimation
Subject RIV: BC - Control Systems Theory
Result website:
http://library.utia.cas.cz/separaty/2008/AS/dedecius-partial%20forgetting%20in%20autoregression%20models.pdf

The assumption of constant parameters of the autoregression model sometimes fails, as the parameters may vary in time. If the parameters vary slowly, the problem is often solved using various forgetting methods like exponential forgetting, linear forgetting etc. However, most of them work on the model parameters probability density function with one common forgetting rate. In the case of different variability of individual parameters, these methods might fail. The developed partial forgetting method gives a new approach, which solves this problem. It releases individual parameters and allows them to change with different rates.

Parciální zapomínání v autoregresních modelech
Permanent Link: http://hdl.handle.net/11104/0163686