A comparison of some new and old robust ridge regression estimators

被引：24

作者：

Ali, Sajid ^{[1
]}

Khan, Himmad ^{[2
]}

Shah, Ismail ^{[1
]}

Butt, Muhammad Moeen ^{[3
]}

Suhail, Muhammad ^{[4
,5
]}

机构：

[1] Quaid I Azam Univ, Dept Stat, Islamabad 45320, Pakistan

[2] Govt Higher Secondary Sch Aboha, Swat, Pakistan

[3] Univ Management & Technol, Sch Business & Econ, Dept Quantitat Methods, Lahore, Pakistan

[4] Univ Punjab, Coll Stat & Actuarial Sci, Lahore, Pakistan

[5] Univ Agr, Dept Stat Math & Comp Sci, Peshawar, Khyber Pakhtunk, Pakistan

来源：

COMMUNICATIONS IN STATISTICS-SIMULATION AND COMPUTATION | 2021年 / 50卷 / 08期

关键词：

Multicollinearity; Ridge regression; Monte Carlo Simulation; BIASED ESTIMATION; SIMULATION;

D O I：

10.1080/03610918.2019.1597119

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：

Ridge regression is used to circumvent the problem of multicollinearity among predictors and many estimators for ridge parameter are available in the literature. However, if the level of collinearity among predictors is high, the existing estimators also have high mean square errors (MSE). In this paper, we consider some existing and propose new estimators for the estimation of ridge parameter k. Extensive Monte Carlo simulations as well as a real-life example are used to evaluate the performance of proposed estimators based on the MSE criterion. The results show the superiority of our proposed estimators compared to the existing estimators.

引用

页码：2213 / 2231

页数：19

共 16 条

[1] [Anonymous], 2012, REGRESSION ANAL EXAM
[2] [Anonymous], 1999, Metron
[3] SIMULATION STUDY OF ALTERNATIVES TO ORDINARY LEAST-SQUARES
DEMPSTER, AP
SCHATZOFF, M
WERMUTH, N
[J]. JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION, 1977, 72 (357) : 77 - 93
[4] Study of partial least squares and ridge regression methods
Firinguetti, Luis
Kibria, Golam
Araya, Rodrigo
[J]. COMMUNICATIONS IN STATISTICS-SIMULATION AND COMPUTATION, 2017, 46 (08) : 6631 - 6644
[5] A SIMULATION STUDY OF SOME RIDGE ESTIMATORS
GIBBONS, DG
[J]. JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION, 1981, 76 (373) : 131 - 139
[6] GENERALIZED CROSS-VALIDATION AS A METHOD FOR CHOOSING A GOOD RIDGE PARAMETER
GOLUB, GH
HEATH, M
WAHBA, G
[J]. TECHNOMETRICS, 1979, 21 (02) : 215 - 223
[7] BIASED ESTIMATION IN REGRESSION - EVALUATION USING MEAN SQUARED ERROR
GUNST, RF
MASON, RL
[J]. JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION, 1977, 72 (359) : 616 - 628
[8] EXPLICIT AND CONSTRAINED GENERALIZED RIDGE ESTIMATION
HEMMERLE, WJ
BRANTLE, TF
[J]. TECHNOMETRICS, 1978, 20 (02) : 109 - 120
[9] CLASS OF BIASED ESTIMATORS IN LINEAR-REGRESSION
HOCKING, RR
SPEED, FM
LYNN, MJ
[J]. TECHNOMETRICS, 1976, 18 (04) : 425 - 437
[10] RIDGE REGRESSION - BIASED ESTIMATION FOR NONORTHOGONAL PROBLEMS
HOERL, AE
KENNARD, RW
[J]. TECHNOMETRICS, 1970, 12 (01) : 55 - &

← 1 2 →