Premium
On the estimation of parameters for frequency domain models
Author(s) -
Ritzi Robert William,
Sorooshian Soroosh,
Gupta Vijai K.
Publication year - 1991
Publication title -
water resources research
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.863
H-Index - 217
eISSN - 1944-7973
pISSN - 0043-1397
DOI - 10.1029/91wr00069
Subject(s) - frequency domain , estimation , mathematics , nonlinear system , variance (accounting) , estimation theory , algorithm , domain (mathematical analysis) , mathematical optimization , computer science , mathematical analysis , engineering , physics , accounting , systems engineering , quantum mechanics , business
The problem of estimating parameters for frequency domain models is considered. Past approaches have most commonly based estimation criteria upon modulus and phase transformations of the model and sample frequency response functions. As an alternative, a complex vector estimation criterion is proposed and is implemented in a nonlinear, Gauss‐Marquardt optimization algorithm. When compared to the previous methods of fitting modulus and phase transformations, the complex vector estimation methodology has less bias and variance and is more robust.