A Comparison Of Robust Regression Techniques For The Estimation Of Weibull Parameters

Douglas R. Shier; Kenneth D. Lawrence

doi:10.1080/03610918408812412

A Comparison Of Robust Regression Techniques For The Estimation Of Weibull Parameters

Douglas R. Shier, Kenneth D. Lawrence

Research output: Contribution to journal › Article › peer-review

11 Scopus citations

Abstract

In this paper several alternative robust regression techniques are compared for estimating parameters of a Weibull distribution. In addition to the usual least squares (Lj) and least absolute deviation (L) methods, a number of one-step reweighting schemes based on the Lj residuals are considered. The results of an extensive series of Monte Carlo simulation experiments demonstrate that the Anscombe reweighting scheme generally produces the best Weibull estimates over the range of sample sizes and parameter valyes studied.

Original language	English (US)
Pages (from-to)	743-750
Number of pages	8
Journal	Communications in Statistics - Simulation and Computation
Volume	13
Issue number	6
DOIs	https://doi.org/10.1080/03610918408812412
State	Published - Jan 1 1984
Externally published	Yes

All Science Journal Classification (ASJC) codes

Statistics and Probability
Modeling and Simulation

Keywords

Monte Carlo
Weibull distribution
iteratively reweighted regression
least absolute deviation
robust

Access to Document

10.1080/03610918408812412

Cite this

@article{e612de4fa11341859584e88b894a8b02,

title = "A Comparison Of Robust Regression Techniques For The Estimation Of Weibull Parameters",

abstract = "In this paper several alternative robust regression techniques are compared for estimating parameters of a Weibull distribution. In addition to the usual least squares (Lj) and least absolute deviation (L) methods, a number of one-step reweighting schemes based on the Lj residuals are considered. The results of an extensive series of Monte Carlo simulation experiments demonstrate that the Anscombe reweighting scheme generally produces the best Weibull estimates over the range of sample sizes and parameter valyes studied.",

keywords = "Monte Carlo, Weibull distribution, iteratively reweighted regression, least absolute deviation, robust",

author = "Shier, {Douglas R.} and Lawrence, {Kenneth D.}",

note = "Funding Information: ACKNOWLEDGMENTS The research of the first author was supported in part by the National Science Foundation Grant #ISP-8Oll45l (EPSCOR.) A1 1 computations were performed on the IBM 3081-K system a t Clemson University.",

year = "1984",

month = jan,

day = "1",

doi = "10.1080/03610918408812412",

language = "English (US)",

volume = "13",

pages = "743--750",

journal = "Communications in Statistics - Simulation and Computation",

issn = "0361-0918",

publisher = "Taylor and Francis Ltd.",

number = "6",

}

TY - JOUR

T1 - A Comparison Of Robust Regression Techniques For The Estimation Of Weibull Parameters

AU - Shier, Douglas R.

AU - Lawrence, Kenneth D.

N1 - Funding Information: ACKNOWLEDGMENTS The research of the first author was supported in part by the National Science Foundation Grant #ISP-8Oll45l (EPSCOR.) A1 1 computations were performed on the IBM 3081-K system a t Clemson University.

PY - 1984/1/1

Y1 - 1984/1/1

N2 - In this paper several alternative robust regression techniques are compared for estimating parameters of a Weibull distribution. In addition to the usual least squares (Lj) and least absolute deviation (L) methods, a number of one-step reweighting schemes based on the Lj residuals are considered. The results of an extensive series of Monte Carlo simulation experiments demonstrate that the Anscombe reweighting scheme generally produces the best Weibull estimates over the range of sample sizes and parameter valyes studied.

AB - In this paper several alternative robust regression techniques are compared for estimating parameters of a Weibull distribution. In addition to the usual least squares (Lj) and least absolute deviation (L) methods, a number of one-step reweighting schemes based on the Lj residuals are considered. The results of an extensive series of Monte Carlo simulation experiments demonstrate that the Anscombe reweighting scheme generally produces the best Weibull estimates over the range of sample sizes and parameter valyes studied.

KW - Monte Carlo

KW - Weibull distribution

KW - iteratively reweighted regression

KW - least absolute deviation

KW - robust

UR - http://www.scopus.com/inward/record.url?scp=0345565922&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0345565922&partnerID=8YFLogxK

U2 - 10.1080/03610918408812412

DO - 10.1080/03610918408812412

M3 - Article

AN - SCOPUS:0345565922

SN - 0361-0918

VL - 13

SP - 743

EP - 750

JO - Communications in Statistics - Simulation and Computation

JF - Communications in Statistics - Simulation and Computation

IS - 6

ER -

A Comparison Of Robust Regression Techniques For The Estimation Of Weibull Parameters

Abstract

All Science Journal Classification (ASJC) codes

Keywords

Access to Document

Other files and links

Fingerprint

Cite this