Logo PTI
Polish Information Processing Society
Logo FedCSIS

Annals of Computer Science and Information Systems, Volume 11

Proceedings of the 2017 Federated Conference on Computer Science and Information Systems

Least Square Method Robustness of Computations: What is not usually considered and taught

DOI: http://dx.doi.org/10.15439/2017F7

Citation: Proceedings of the 2017 Federated Conference on Computer Science and Information Systems, M. Ganzha, L. Maciaszek, M. Paprzycki (eds). ACSIS, Vol. 11, pages 537–541 ()

Full text

Abstract. There are many practical applications based on the Least Square Error (LSE) method approximation. It is based on a square error minimization ``on a vertical'' axis. The LSE method is simple and easy also for analytical purposes. However, if data span is large over several magnitudes or non-linear LSM is used, severe numerical instability can be expected. The presented contribution describes a simple method for large span of data LSE computation. It is especially convenient if large span of data are to be processed, when the ``standard'' pseudoinverse matrix is ill conditioned. It is actually based on a LSE solution using orthogonal basis vectors instead of orthonormal basis vectors. The presented approach has been used for a linear regression as well as for approximation using radial basis functions.

References

  1. Abatzoglou,T., Mendel,J. 1987. Constrained total least squares, IEEE Conf. Acoust., Speech, Signal Process. (ICASSP’87), Vol. 12, 1485–1488.
  2. Alciatore,D., Miranda,R. 1995. The best least-square line fit, Graphics Gems V (Ed.Paeth,A.W.), 91-97, Academic Press.
  3. Amiri-Simkooei,A.R.,Jazaeri,S. 2012.Weighted total least squares formulated by standard least squares theory,J.Geodetic Sci.,2(2):113-124.
  4. Charpa,S., Canale,R. 1988. Numerical methods for Engineers, McGraw-Hill.
  5. Chatfield,C. 1970. Statistics for technology, Penguin Book.
  6. de Groen,P. 1996 An introduction to total least squares, Nieuw Archief voor Wiskunde,Vierde serie, deel 14, 237–253
  7. DeGroat,R.D., Dowling,E.M. 1993 The data least squares problem and channel equalization. IEEE Trans. Signal Processing, Vol. 41(1), 407–411.
  8. Golub,G.H., Van Loan,C.F. 1980. An analysis of the total least squares problem. SIAM J. on Numer. Anal., 17, 883–893.
  9. Jo,S., Kim,S.W. 2005. Consistent normalized least mean square filtering with noisy data matrix. IEEE Trans. Signal Proc.,Vol. 53(6), 2112–2123.
  10. Kryszig,V. 1983. Advanced engineering mathematics, John Wiley & Sons.
  11. Lee,S.L. 1994. A note on the total least square fit to coplanar points, Tech.Rep. ORNL-TM-12852, Oak Ridge National Laboratory.
  12. Levy,D., 2010. Introduction to Numerical Analysis, Univ.of Maryland
  13. Nievergelt,Y. 1994. Total least squares: State of the Art regression in numerical mathematics, SIAM Review, Vol.36, 258-264
  14. Nixon,M.S., Aguado,A.S. 2012. Feature extraction & image processing for computer vision, Academic Press.
  15. Markowsky,I., VanHueffel,S., 2007. Overview of total least square methods, Signal Processing, 87 (10), 2283-2302.
  16. Pearson,K., 1901. On line and planes of closest fit to system of points in space, Phil.Mag., Vol.2, 559-572
  17. Skala,V., 2016.Total Least Square Error Computation in E2: A New Simple, Fast and Robust Algorithm, CGI 2016 Proc., ACM, pp.1-4, Greece
  18. Skala,V., 2016. A new formulation for total Least Square Error method in d-dimensional space with mapping to a parametric line, ICNAAM 2015, AIP Conf. Proc.1738, pp.480106-1 - 480106-4, Greece
  19. Skala,V., 2017. Least Square Error method approximation and extended cross product using projective representation, ICNAAM 2016 conf., to appear in ICNAAM 2016 proceedings, AIP Press
  20. Skala,V., 2008. Barycentric Coordinates Computation in Homogeneous Coordinates, Computers & Graphics, Elsevier, , Vol.32, No.1, pp.120-127
  21. Van Huffel,S., Vandewalle,J 1991. The total least squares problems: computational aspects and analysis. SIAM Publications, Philadelphia PA.
  22. van Huffel,S., Lemmerling,P. 2002. Total least squares and errors-in-variables modeling: Analysis, algorithms and applications. Dordrecht, The Netherlands: Kluwer Academic Publishers
  23. Perpendicular regression of a line (download 2015-12-20) http://www.mathpages.com/home/kmath110.htm
  24. Skala,V., 2017. RBF Interpolation with CSRBF of Large Data Sets, ICCS 2017, Procedia Computer Science, Vol.108, pp. 2433-2437, Elsevier
  25. Skala,V., 2017. High Dimensional and Large Span Data Least Square Error: Numerical Stability and Conditionality, accepted to ICPAM 2017