Research Article Open Access

Linear Regression Model Selection Based on Robust Bootstrapping Technique

Hassan S. Uraibi1, Habshah Midi1, Bashar A. Talib1 and Jabar H. Yousif1
  • 1 ,
American Journal of Applied Sciences
Volume 6 No. 6, 2009, 1191-1198

DOI: https://doi.org/10.3844/ajassp.2009.1191.1198

Submitted On: 9 January 2009 Published On: 30 June 2009

How to Cite: Uraibi, H. S., Midi, H., Talib, B. A. & Yousif, J. H. (2009). Linear Regression Model Selection Based on Robust Bootstrapping Technique. American Journal of Applied Sciences, 6(6), 1191-1198. https://doi.org/10.3844/ajassp.2009.1191.1198

Abstract

Problem statement: Bootstrap approach had introduced new advancement in modeling and model evaluation. It was a computer intensive method that can replace theoretical formulation with extensive use of computer. The Ordinary Least Squares (OLS) method often used to estimate the parameters of the regression models in the bootstrap procedure. Unfortunately, many statistics practitioners are not aware of the fact that the OLS method can be adversely affected by the existence of outliers. As an alternative, a robust method was put forward to overcome this problem. The existence of outliers in the original sample may create problem to the classical bootstrapping estimates. There was possibility that the bootstrap samples may contain more outliers than the original dataset, since the bootstrap re-sampling is with replacement. Consequently, the outliers will have an unduly effect on the classical bootstrap mean and standard deviation. Approach: In this study, we proposed to use a robust bootstrapping method which was less sensitive to outliers. In the robust bootstrapping procedure, we proposed to replace the classical bootstrap mean and standard deviation with robust location and robust scale estimates. A number of numerical examples were carried out to assess the performance of the proposed method. Results: The results suggested that the robust bootstrap method was more efficient than the classical bootstrap. Conclusion/Recommendations: In the presence of outliers in the dataset, we recommend using the robust bootstrap procedure as its estimates are more reliable.

  • 1,682 Views
  • 3,507 Downloads
  • 8 Citations

Download

Keywords

  • Bootstrap
  • outliers
  • robust location
  • robust standard deviation