Skip to main content
Log in

Algorithms for the Huber estimator in multiple regression

Algorithmen für den Huber-Schätzer in der multiplen Regression

  • Published:
Computing Aims and scope Submit manuscript

Abstract

Several iterative procedures have been proposed and developed to solve numerically the problem of robust regression, in particular, of robust linear regression. The algorithms described here are modified versions of the “sophisticated method” given by Huber (1973, [8]) which sometimes fail to converge. In this paper, the new algorithms are formulated and convergence proofs are given. The behavior of the procedures is illustrated by a numerical example and is compared to another (“simple”) algorithm.

Zusammenfassung

Verschiedene iterative Prozeduren zur numerischen Lösung des Problems der robusten Regression, genauer der robusten linearen Regression, wurden vorgeschlagen und entwickelt. Die hier beschriebenen Algorithmen sind modifizierte Fassungen der in der Veröffentlichung von Huber (1973, [8]) beschriebenen raffinierten („sophisticated”) Methode, die aber manchmal nicht konvergiert. In der vorliegenden Arbeit werden neue Algorithmen formuliert und Konvergenzbeweise gegeben. Das Verhalten der Prozeduren wird an einem numerischen Beispiel illustriert und mit dem eines einfachen („simple”) Algorithmus verglichen.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Similar content being viewed by others

References

  1. Andrews, D. F.: A Robust Method for Multiple Linear Regression. Technometrics16, 523–531 (1974).

    Google Scholar 

  2. Brownlee, K. A.: Statistical Theory and Methodology in Science and Engineering, 2nd edition, Section 13.12. New York: Wiley 1965.

    Google Scholar 

  3. Daniel, C., Wood, F. S.: Fitting Equations to Data, Chapter 5. New York: Wiley 1971.

    Google Scholar 

  4. Dutter, R.: Robust Regression: Different Approaches to Numerical, Solutions and Algorithms. Research Report No. 6, Fachgruppe für Statistik, ETH, Zurich (1975a).

    Google Scholar 

  5. Dutter, R.: Numerical Solution of Robust Regression Problems: Computational Aspects, a Comparison. Research Report No. 7, Fachgruppe für Statistik, ETH, Zurich (1975b).

    Google Scholar 

  6. Ekblom, H.: L p -Methods for Robust Regression. Bit.14, 22–32 (1974).

    Google Scholar 

  7. Huber, P. J.: Robust Estimation of a Location Parameter. Ann. Math. Statist.35, 73–101 (1964).

    Google Scholar 

  8. Huber, P. J.: Robust Regression: Asymptotics, Conjectures and Monte Carlo. Ann. Statist.1, 799–821 (1973).

    Google Scholar 

  9. Huber, P. J., Dutter, R.: Numerical Solution of Robust Regression Problems. COMPSTAT 1974, Proc. Comp. Statist. (Bruckmann, G., ed.). Wien: Physika Verlag 1974.

    Google Scholar 

  10. Rey, W.: On the Leastp-th Power Methods in Multiple Regressions and Location Estimation. Bit15, 2, 174–184 (1975).

    Google Scholar 

  11. Dutter, R., Huber, P. J.: On Methods for the Numerical Solution of Robust Regression Problems. Submitted for publication (1976).

Download references

Author information

Authors and Affiliations

Authors

Additional information

Work supported by the Swiss National Science Foundation, Contract No. 2.836.73.

Rights and permissions

Reprints and permissions

About this article

Cite this article

Dutter, R. Algorithms for the Huber estimator in multiple regression. Computing 18, 167–176 (1977). https://doi.org/10.1007/BF02243626

Download citation

  • Received:

  • Revised:

  • Issue Date:

  • DOI: https://doi.org/10.1007/BF02243626

Keywords

Navigation