### Abstract

Least squares fitting of regression models is a widely used technique. The presence of outliers in the data can have an adverse effect on the method of least squares, resulting in a model that does not adequately fit to the bulk of the data. For this situation, robust regression techniques have been proposed as an improvement to the method of least squares. We propose a robust regression procedure that performs well relative to the current robust methods against a variety of dataset types. Evaluations are performed using datasets without outliers (testing efficiency), with a large percentage of outliers (testing breakdown), and with high leverage outliers (testing bounded influence). The datasets are based on 2-level factorial designs that include axial points to evaluate leverage effects. A Monte Carlo simulation approach is used to evaluate the estimating capability of the proposed procedure relative to several competing methods. We also provide an application to estimating costs for government satellites.

Original language | English (US) |
---|---|

Pages (from-to) | 125-139 |

Number of pages | 15 |

Journal | Naval Research Logistics |

Volume | 45 |

Issue number | 2 |

State | Published - Mar 1998 |

### Fingerprint

### ASJC Scopus subject areas

- Management Science and Operations Research

### Cite this

*Naval Research Logistics*,

*45*(2), 125-139.

**A robust regression technique using compound estimation.** / Simpson, James R.; Montgomery, Douglas.

Research output: Contribution to journal › Article

*Naval Research Logistics*, vol. 45, no. 2, pp. 125-139.

}

TY - JOUR

T1 - A robust regression technique using compound estimation

AU - Simpson, James R.

AU - Montgomery, Douglas

PY - 1998/3

Y1 - 1998/3

N2 - Least squares fitting of regression models is a widely used technique. The presence of outliers in the data can have an adverse effect on the method of least squares, resulting in a model that does not adequately fit to the bulk of the data. For this situation, robust regression techniques have been proposed as an improvement to the method of least squares. We propose a robust regression procedure that performs well relative to the current robust methods against a variety of dataset types. Evaluations are performed using datasets without outliers (testing efficiency), with a large percentage of outliers (testing breakdown), and with high leverage outliers (testing bounded influence). The datasets are based on 2-level factorial designs that include axial points to evaluate leverage effects. A Monte Carlo simulation approach is used to evaluate the estimating capability of the proposed procedure relative to several competing methods. We also provide an application to estimating costs for government satellites.

AB - Least squares fitting of regression models is a widely used technique. The presence of outliers in the data can have an adverse effect on the method of least squares, resulting in a model that does not adequately fit to the bulk of the data. For this situation, robust regression techniques have been proposed as an improvement to the method of least squares. We propose a robust regression procedure that performs well relative to the current robust methods against a variety of dataset types. Evaluations are performed using datasets without outliers (testing efficiency), with a large percentage of outliers (testing breakdown), and with high leverage outliers (testing bounded influence). The datasets are based on 2-level factorial designs that include axial points to evaluate leverage effects. A Monte Carlo simulation approach is used to evaluate the estimating capability of the proposed procedure relative to several competing methods. We also provide an application to estimating costs for government satellites.

UR - http://www.scopus.com/inward/record.url?scp=0032024053&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0032024053&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:0032024053

VL - 45

SP - 125

EP - 139

JO - Naval Research Logistics

JF - Naval Research Logistics

SN - 0894-069X

IS - 2

ER -