×

Some properties of generalized ridge estimators. (English) Zbl 0555.62058

Generalized ridge estimators such as the ones proposed by A. E. Hoerl and R. W. Kennard [Technometrics 12, 55-67 and 69-82 (1970; Zbl 0202.172)] have been examined by the authors for comparing analytically, rather than by Monte-Carlo methods, with the usual least squares estimators. This is done by finding the first two moments of the estimators in the form of an infinite series and evaluating them approximatively by recursive formulas. The relative mean-square error, bias and efficiency with respect to the usual estimators are all presented.
Reviewer: A.M.Kshirsagar

MSC:

62J07 Ridge regression; shrinkage estimators (Lasso)

Citations:

Zbl 0202.172
PDFBibTeX XMLCite
Full Text: DOI

References:

[1] Davis P. J., Handbook of Mathematical Functions pp 253– (1964)
[2] Draper N. R., Techonometrics 21 pp 451– (1979) · doi:10.1080/00401706.1979.10489815
[3] Dwivedi T. W., Technometrics 22 pp 205– (1980) · doi:10.1080/00401706.1980.10486136
[4] Hemmerle W. J., Technometrics 17 pp 309– (1975) · doi:10.1080/00401706.1975.10489333
[5] Hemmerle W. J., Technometrics 20 pp 109– (1978) · doi:10.1080/00401706.1978.10489634
[6] Hocking R. R., Technometrics 18 pp 425– (1976) · doi:10.1080/00401706.1976.10489474
[7] Hoerl A. E., Technometrics 12 pp 55– (1970) · doi:10.1080/00401706.1970.10488634
[8] Lawless J. R., Communications in Statistics 7 pp 139– (1978) · Zbl 0375.62067 · doi:10.1080/03610927808827609
[9] Obenchain R. L., Technometrics 17 pp 431– (1975) · doi:10.1080/00401706.1975.10489369
[10] Resnikoff G. J., Tables of the Non-Central t-Distribution (1957) · Zbl 0077.12003
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. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.