On identification methods of linear regression models

Abstract

The aim of this paper is to discus the efficiency of the least squares and least absolute values estimation methods, in such situations, then the sample is with outliers, the variance of the residuals changes in the time and the residuals are correlated. A range of different modifications of these methods in order to improve accuracy of the estimation is offend. All the discussed procedures are fulfilled in a PC by means of SAS (Statistical Analysis System).