Journal of Geodesy

, Volume 82, Issue 7, pp 415–421 | Cite as

On weighted total least-squares adjustment for linear regression

Original Article

Abstract

The weighted total least-squares solution (WTLSS) is presented for an errors-in-variables model with fairly general variance–covariance matrices. In particular, the observations can be heteroscedastic and correlated, but the variance–covariance matrix of the dependent variables needs to have a certain block structure. An algorithm for the computation of the WTLSS is presented and applied to a straight-line fit problem where the data have been observed with different precision, and to a multiple regression problem from recently published climate change research.

Keywords

Total least-squares solution (TLSS) Errors-in-variables model Weight matrix Heteroscedastic observations Straight-line fit Multiple linear regression 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Golub GH, van Loan Ch (1980) An analysis of the total least-squares problem. SIAM J Num Anal 17: 883–893CrossRefGoogle Scholar
  2. 2.
    Golub GH, van Loan Ch (1996) Matrix computations, 3rd edn. Johns Hopkins University Press, BaltimoreGoogle Scholar
  3. 3.
    Grafarend E, Schaffrin B (1993) Adjustment computations in linear models (in German). Bibliographical Institute, Mannheim, GermanyGoogle Scholar
  4. 4.
    Magnus JR, Neudecker H (1988) Matrix differential calculus with applications in statistics and econometrics. Wiley, New YorkGoogle Scholar
  5. 5.
    Mann ME, Emanuel KA (2006) Atlantic hurricane trends linked to climate change. EOS 87–24: 233–241Google Scholar
  6. 6.
    Markovsky I, Rastello M, Premoli A, Kukush A, van Huffel S (2006) The element-wise weighted total least-squares problem. Comput Statist Data Anal 50: 181–209CrossRefGoogle Scholar
  7. 7.
    Neri F, Saitta G, Chiofalo S (1989) An accurate and straightforward approach to line regression analysis of error-affected experimental data. J Phys Ser E: Sci Instr 22: 215–217CrossRefGoogle Scholar
  8. 8.
    Pope AJ (1972) Some pitfalls to be avoided in the iterative adjustment of nonlinear problems. In: Proc 38th Ann Meet Am Soc Phot, Washington, DC, pp 449–473Google Scholar
  9. 9.
    Schaffrin B, Wang Z (1995) Multiplicative outlier search using homBLUP and an equivalence theorem. Manus Geodaet 20: 21–26Google Scholar
  10. 10.
    Schaffrin B, Lee IP, Felus Y, Choi YS (2006) Total least-squares for geodetic straight-line and plane adjustment. Boll Geod Sci Aff 65: 141–168Google Scholar
  11. 11.
    Schaffrin B, Felus Y (2007) Multivariate total-least squares adjustment for empirical affine transformations. In: Proc 6th Hotine-Marussi Symp, Wuhan, People’s Republic of China, May, forthcomingGoogle Scholar
  12. 12.
    Schuermans M, Markovsky I, van Huffel S (2007) An adapted version of the element-wise weighted total least-squares method for applications in cheomometrics. Chemometrics Intell Lab Syst 85: 40–46CrossRefGoogle Scholar
  13. 13.
    van Huffel S, Vandewalle J (1991) The total least-squares problem. Computational aspects and analysis. SIAM, PhiladelphiaGoogle Scholar
  14. 14.
    van Huffel S, Vandewalle J (1989) Analysis and properties of the generalized total least-squares problem \({\rm AX} \approx {\rm B}\) when some or all columns of A are subject to error. SIAM J Matrix Anal Appl 10: 294–315CrossRefGoogle Scholar
  15. 15.
    Xu P (1999) Despeckling SAR-type multiplicative noise. Int J Remote Sensing 20: 2577–2596CrossRefGoogle Scholar
  16. 16.
    Xu P, Shimada S (2000) Least-squares parameter estimation in multiplicative noise models. Commun Statist B 29: 83–96CrossRefGoogle Scholar

Copyright information

© Springer-Verlag 2007

Authors and Affiliations

  1. 1.Geodetic Science ProgramThe Ohio State UniversityColumbusUSA
  2. 2.Engineering Geodesy and Measurement SystemsGraz University of TechnologyGrazAustria

Personalised recommendations