TY - JOUR
T1 - Tracking time-varying parameters with local regression
AU - Joensen, Alfred Karsten
AU - Nielsen, Henrik Aalborg
AU - Nielsen, Torben Skov
AU - Madsen, Henrik
PY - 2000
Y1 - 2000
N2 - This paper shows that the recursive least-squares (RLS) algorithm with forgetting factor is a special case of a varying-coe\$cient model, and a model which can easily be estimated via simple local regression. This observation allows us to formulate a new method which retains the RLS algorithm, but extends the algorithm by including polynomial approximations. Simulation results are provided, which indicates that this new method is superior to the classical RLS method, if the parameter variations are smooth.
AB - This paper shows that the recursive least-squares (RLS) algorithm with forgetting factor is a special case of a varying-coe\$cient model, and a model which can easily be estimated via simple local regression. This observation allows us to formulate a new method which retains the RLS algorithm, but extends the algorithm by including polynomial approximations. Simulation results are provided, which indicates that this new method is superior to the classical RLS method, if the parameter variations are smooth.
KW - Recursive estimation; Varying-coe\$cient; Conditional parametric; Polynomial approximation; Weighting functions.
U2 - 10.1016/S0005-1098(00)00029-7
DO - 10.1016/S0005-1098(00)00029-7
M3 - Journal article
SN - 0005-1098
VL - 36
SP - 1199
EP - 1204
JO - Automatica
JF - Automatica
IS - 8
ER -