Skip to search boxSkip to navigationSkip to main content

Monte-Carlo study of some robust estimators: the simple linear regression case

  • Tai Solarin University of Education
    ,
  • Federal University of Technology, Akure
    ,
  • Richmond, The American International University in London
Research Output: Contribution to journal Article Peer-review

Open access

Abstract

In this study, Least Trimmed Squares (LTS), Theil’s Pair-wise Median (Theil) and Bayesian estimation methods (BAYES) are compared relative to the OLSE via Monte-Carlo Simulation. Variance, Bias, Mean Square Error (MSE) and Relative Mean Square Error (RMSE) were calculated to evaluate the estimators’ performance. The Simple Linear Regression model is explored for the conditions in which the error term is assumed to be drawn from three error distributions: unit normal, lognormal and Cauchy. Theil’s non-parametric estimation procedure was found to have the strongest and most reliable performance. The subsequent-best results are acquired from LTS approach Though it was observed that the Bayesian estimators are affected by deviation of the dataset from normality, yet it is established from the results that the Bayesian estimators performed optimally more than all other competitors, even under non normal situations (especially under the standard lognormal distribution) in some cases, except whenever the error is drawn from a heavy tail distribution (Lognormal and Cauchy)..OLSE is most effective reliable as long as the normality assumptions preserve

Publication Information

Output type

Research Output: Contribution to journal Article Peer-review

Original language

English

Article number

6

Pages from-to (Number of pages)

Pages 1-11

Journal (Volume, Issue Number)

University of Wah Journal of Science and Technology (Volume 8)

Publication milestones

  • Accepted/In press - 19/11/2024
  • Published - 30/12/2024

Publication status

Published - 30/12/2024

ISSN

2523-0123

External Publication IDs

  • handle.net: 10547/626615

Access to documents

Publication metrics

Metrics

Download statistics
Download count
4