Title :
Parameter Estimations in Linear Regression Models with AR(2) Errors in Which the Parameters Have a Special Relationship
Author :
Xu, Wenke ; Liu, Fuxiang ; Li, Fengri ; Wu, Haijun ; Jin, Xuejing
Author_Institution :
Northeast Forestry Univ., Harbin, China
Abstract :
The purpose of this paper is to study parameters estimations in linear regression model with AR(2) errors ¿t = ¿1¿t-1 + ¿2¿t-2 - ¿t, t = 1, 2,¿, n in which the parameters have a special relationship ¿2 = ¿1 2. For the properties of variance-covariance matrix ¿ , This kind of models are transformed into the standard linear regression models without autocorrelation errors and apply the method of cycle generalized least squares (CGLS) to estimate parameters. Simulation results show that efficiency of CGLS method is superior over the method of generalized least squares (GLS) under mean square error criterion.
Keywords :
parameter estimation; regression analysis; autocorrelation errors; cycle generalized least squares; linear regression model; mean square error criterion; parameter estimation; variance-covariance matrix; Autocorrelation; Covariance matrix; Equations; Gaussian processes; Least squares approximation; Least squares methods; Linear regression; Mean square error methods; Parameter estimation; Vectors;
Conference_Titel :
Computational Intelligence and Software Engineering, 2009. CiSE 2009. International Conference on
Conference_Location :
Wuhan
Print_ISBN :
978-1-4244-4507-3
Electronic_ISBN :
978-1-4244-4507-3
DOI :
10.1109/CISE.2009.5364199