A General Class of Estimators for the Linear Regression Model Affected by Collinearity and Outliers

被引:1
作者
Macedo, Pedro [1 ]
Scotto, Manuel [1 ]
Silva, Elvira [2 ]
机构
[1] Univ Aveiro, Dept Matemat, P-3810193 Aveiro, Portugal
[2] Univ Porto, Fac Econ, P-4100 Oporto, Portugal
关键词
Collinearity; Linear regression; Maximum entropy; Outliers; Robust regression; MAXIMUM-ENTROPY APPROACH; STATISTICAL MECHANICS; INFORMATION THEORY; RIDGE REGRESSION; ECONOMETRICS;
D O I
10.1080/03610911003695719
中图分类号
O21 [概率论与数理统计]; C8 [统计学];
学科分类号
020208 ; 070103 ; 0714 ;
摘要
In this article, a general class of estimators for the linear regression model affected by outliers and collinearity is introduced and studied in some detail. This class of estimators combines the theory of light, maximum entropy, and robust regression techniques. Our theoretical findings are illustrated through a Monte Carlo simulation study.
引用
收藏
页码:981 / 993
页数:13
相关论文
共 42 条
[21]   ON INFORMATION AND SUFFICIENCY [J].
KULLBACK, S ;
LEIBLER, RA .
ANNALS OF MATHEMATICAL STATISTICS, 1951, 22 (01) :79-86
[22]  
Kullback S., 1959, INFORM THEORY STAT
[23]   Estimation of multi-output production functions with incomplete data: A generalised maximum entropy approach [J].
Lence, SH ;
Miller, DJ .
EUROPEAN REVIEW OF AGRICULTURAL ECONOMICS, 1998, 25 (02) :188-209
[24]  
Levine R. D., 1979, MAXIMUM ENTROPY FORM
[25]   ON A MEASURE OF THE INFORMATION PROVIDED BY AN EXPERIMENT [J].
LINDLEY, DV .
ANNALS OF MATHEMATICAL STATISTICS, 1956, 27 (04) :986-1005
[26]   Using Liu-type estimator to combat collinearity [J].
Liu, KJ .
COMMUNICATIONS IN STATISTICS-THEORY AND METHODS, 2003, 32 (05) :1009-1020
[27]  
LIU KJ, 1993, COMMUN STAT THEORY, V22, P393
[28]  
Maronna R.A., 2018, ROBUST STAT THEORY M
[29]   Probability density estimation using entropy maximization [J].
Miller, G ;
Horn, D .
NEURAL COMPUTATION, 1998, 10 (07) :1925-1938
[30]  
Mishra S.K., 2004, Economics Bulletin, V3, P1