Robust regression for estimating the burr xii parameters with outliers

Yung Fu Cheng, Fu Kwun Wang

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

The Burr XII distribution offers an even more flexible alternative to the lognormal and the log-logistic distributions. This paper presents a study of the efficiency of using robust regression, M-estimate method, to estimate the parameters of two-parameter Burr XII distribution. Our proposed method will compare with other methods such as unweighted least squares method and maximum likelihood estimation method. The simulated data for complete data and multiple-censoring data with outliers were used to validate the comparison study. Results show that the robust regression method outperforms other methods in most cases in terms of bias and root mean square error.

Original languageEnglish
Title of host publication38th International Conference on Computers and Industrial Engineering 2008
Pages338-347
Number of pages10
StatePublished - 2008
Externally publishedYes
Event38th International Conference on Computers and Industrial Engineering 2008 - Beijing, China
Duration: 31 10 200802 11 2008

Publication series

Name38th International Conference on Computers and Industrial Engineering 2008
Volume1

Conference

Conference38th International Conference on Computers and Industrial Engineering 2008
Country/TerritoryChina
CityBeijing
Period31/10/0802/11/08

Keywords

  • Burr xii distribution
  • Least squares estimator
  • Maximum likelihood estimator
  • Mestimator
  • Robust regression estimator

Fingerprint

Dive into the research topics of 'Robust regression for estimating the burr xii parameters with outliers'. Together they form a unique fingerprint.

Cite this