This paper deals with the regressi on optimality of combining ridge and prinapal components estimator.It is proved that the combining ridge and principal compenents estimator of regression coefficient has several type...This paper deals with the regressi on optimality of combining ridge and prinapal components estimator.It is proved that the combining ridge and principal compenents estimator of regression coefficient has several types of minimum—Variance preperties.Some results of the paper[2] are the special cases of this paper.We also compare the combining ridge and principal components estimator with the principal components esti- mator in several senses.展开更多
In this paper,we propose a new biased estimator of the regression parameters,the generalized ridge and principal correlation estimator.We present its some properties and prove that it is superior to LSE(least squares ...In this paper,we propose a new biased estimator of the regression parameters,the generalized ridge and principal correlation estimator.We present its some properties and prove that it is superior to LSE(least squares estimator),principal correlation estimator,ridge and principal correlation estimator under MSE(mean squares error) and PMC(Pitman closeness) criterion,respectively.展开更多
文摘This paper deals with the regressi on optimality of combining ridge and prinapal components estimator.It is proved that the combining ridge and principal compenents estimator of regression coefficient has several types of minimum—Variance preperties.Some results of the paper[2] are the special cases of this paper.We also compare the combining ridge and principal components estimator with the principal components esti- mator in several senses.
基金Foundation item： the National Natural Science Foundation of China （Nos. 60736047 10671007+2 种基金 60772036） the Foundation of Beijing Jiaotong University （Nos. 2006XM037 2007XM046）.
文摘In this paper,we propose a new biased estimator of the regression parameters,the generalized ridge and principal correlation estimator.We present its some properties and prove that it is superior to LSE(least squares estimator),principal correlation estimator,ridge and principal correlation estimator under MSE(mean squares error) and PMC(Pitman closeness) criterion,respectively.
