IDENTIFICATION PROPERTIES ENHANCEMENT ALGORITHM FOR PROBLEMS OF PARAMETERS ESTIMATION OF LINEAR REGRESSION MODEL
This brief paper describes a new approach to identification of unknown constant parameters for a linear regression model. The main idea of the method lies in transformation of initial model into a new kind one. The new model regressor possesses identification properties or meets persistency of excit...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Saint Petersburg National Research University of Information Technologies, Mechanics and Optics (ITMO University)
2016-05-01
|
| Series: | Naučno-tehničeskij Vestnik Informacionnyh Tehnologij, Mehaniki i Optiki |
| Subjects: |
| Summary: | This brief paper describes a new approach to identification of unknown constant parameters for a linear regression model. The main idea of the method lies in transformation of initial model into a new kind one. The new model regressor possesses identification properties or meets persistency of excitation conditions. An example of two unknown parameters identification for the linear regression model shows efficiency of the proposed approach. Simulation was carried out for a regressor with no persistency of excitation conditions, hence, parameter identification is not guaranteed. |
|---|---|
| ISSN: | 2226-1494 2500-0373 |