Using Liu-type estimator to combat collinearity. (English) Zbl 1107.62345

Summary: Linear regression model and least squares method are widely used in many fields of natural and social sciences. In the presence of collinearity, the least squares estimator is unstable and often gives misleading information. Ridge regression is the most common method to overcome this problem. We find that when there exists severe collinearity, the shrinkage parameter selected by existing methods for ridge regression may not fully address the ill conditioning problem. To solve this problem, we propose a new two-parameter estimator. We show using both theoretic results and simulation that our new estimator has two advantages over ridge regression. First, our estimator has less mean squared error (MSE). Second, our estimator can fully address the ill conditioning problem. A numerical example from literature is used to illustrate the results.


62J07 Ridge regression; shrinkage estimators (Lasso)
62F10 Point estimation
Full Text: DOI


[1] DOI: 10.1080/03610929508831585 · Zbl 0937.62612
[2] DOI: 10.1081/STA-100104357 · Zbl 1008.62632
[3] Akdeniz F., Indian Journal of Statistics 63 pp 321– (2001)
[4] DOI: 10.1080/02664760021817 · Zbl 0970.62015
[5] DOI: 10.1002/0471725153
[6] DOI: 10.1080/01621459.1981.10477619
[7] Gruber M. H. J., Improving Efficiency by Shrinkage: The James-Stein and Ridge Regression Estimators (1998) · Zbl 0920.62085
[8] Hald A., Statistical Theory with Engineering Applications (1952) · Zbl 0048.36404
[9] DOI: 10.1080/00401706.1978.10489634
[10] DOI: 10.1080/00401706.1970.10488634
[11] DOI: 10.1080/03610927608827333 · Zbl 0326.62047
[12] DOI: 10.1080/03610927508827232 · Zbl 0296.62062
[13] Kaciranlar S., Indian Journal of Statistics 61 pp 443– (1999)
[14] DOI: 10.1081/STA-100108454 · Zbl 1009.62560
[15] DOI: 10.1080/03610929308831027 · Zbl 0784.62065
[16] DOI: 10.1080/01621459.1975.10479882
[17] DOI: 10.1081/STA-100002036 · Zbl 1009.62559
[18] DOI: 10.1080/00401706.1978.10489675
[19] DOI: 10.1021/ie50275a002
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. It attempts to reflect the references listed in the original paper as accurately as possible without claiming the completeness or perfect precision of the matching.