Generalized Shrunken Type GM Estimation in Linear Model

摘要

The parameter estimation problem in linear model is considered when multicollinearity and outliers exist simultaneously.A class of new robust biased estimator,Generalized Shrunken Type-GM Estimation,with their calculated methods are established by combine of GM estimator and biased estimator include Ridge estimate,Principal components estimate and Liu estimate and so on.A numerical example shows that the most attractive advantage of these new estimators is that they can not only overcome the multicollinearity of coefficient matrix and outliers but also have the ability to control the influence of leverage points.