MULTISTAGE PARAMETER ESTIMATION PROCEDURE FOR CONTINUOUS SYSTEMS.
- Additional Document Info
- View All
The discrete-time least squares approach is extended to estimate the parameters in nonlinear continuous systems. The resulting direct integral least squares (DILS) method is simple and numerically efficient. According to the theory of biased estimators the DILS estimation may be preferable to the conventional (indirect) LS method even from a statistical point of view. If necessary, biasedness of the DILS estimates are removed by continuous-time instrumental variable algorithms. A four-stage estimation procedure is presented which preserves the advantages of its stages, thereby combining robustness and efficiency of the direct method with unbiasedness of the indirect estimates.
author list (cited authors)
Vajda, S., Valko, P., & Yermakova, A
complete list of authors
Vajda, S||Valko, P||Yermakova, A