Partial Forgetting. A new method for tracking time-variant parameters

0322746 - ÚTIA 2009 CZ eng V - Výzkumná zpráva
Dedecius, Kamil - Nagy, Ivan - Kárný, Miroslav - Pavelková, Lenka
Partial Forgetting. A new method for tracking time-variant parameters.
[Parciální zapomínání. Nová metoda sledování časově proměnných parametrů.]
Praha: UTIA AV ČR, 2009. 20 s. Research Report, 2249.
Grant CEP: GA MŠMT 2C06001
Výzkumný záměr: CEZ:AV0Z10750506
Klíčová slova: parameter estimation * forgetting * modelling
Kód oboru RIV: BD - Teorie informace
http://library.utia.cas.cz/separaty/2009/AS/dedecius-partial forgetting. a new method for tracking time-variant parameters.pdf

Tracking of slowly varying parameters is an important task in the theory of adaptive systems. Majority of prediction and control algorithms, employing regression models like autoregression model (AR), autoregression model with exogenous inputs (ARX), autoregression model with moving average (ARMA) etc., assume a carefully defined model structure and correctly estimated parameters. Problems arise, when the model parameters vary in time. The problems of slowly time-varying model parameters were given a thorough attention. The proposed partial forgetting method tries to solve this issue by a new approach.
Trvalý link: http://hdl.handle.net/11104/0170912