Statistical analysis of curve fitting methods in errors-in-variables models

Al-Sharadqah, A.; Chernov, N.

doi:10.1090/S0094-9000-2012-00860-0

Statistical analysis of curve fitting methods in errors-in-variables models

Authors: A. Al-Sharadqah and N. Chernov
Journal: Theor. Probability and Math. Statist. 84 (2012), 1-14
MSC (2010): Primary 62H10, 62J02; Secondary 62H35
DOI: https://doi.org/10.1090/S0094-9000-2012-00860-0
Published electronically: July 26, 2012
MathSciNet review: 2857412
Full-text PDF Free Access

Abstract | References | Similar Articles | Additional Information

Abstract: Regression models in which all variables are subject to errors are known as errors-in-variables (EIV) models. The respective parameter estimates have many unusual properties: their exact distributions are very hard to determine, and their absolute moments are often infinite (so that their mean and variance do not exist). In our paper, Error analysis for circle fitting algorithms, Electr. J. Stat. 3 (2009), 886–911, we developed an unconventional statistical analysis that allowed us to effectively assess EIV parameter estimates and design new methods with superior characteristics. In this paper we validate our approach in a series of numerical tests. We also prove that in the case of fitting circles, the estimates of the parameters are absolutely continuous (have densities).

References

R. J. Adcock, Note on the method of least squares, Analyst 4 (1877), 183–184.

Ali Al-Sharadqah and Nikolai Chernov, Error analysis for circle fitting algorithms, Electron. J. Stat. 3 (2009), 886–911. MR 2540845, DOI https://doi.org/10.1214/09-EJS419

Yasuo Amemiya and Wayne A. Fuller, Estimation for the nonlinear functional relationship, Ann. Statist. 16 (1988), no. 1, 147–160. MR 924862, DOI https://doi.org/10.1214/aos/1176350696

T. W. Anderson, Estimation of linear functional relationships: approximate distributions and connections with simultaneous equations in econometrics, J. Roy. Statist. Soc. Ser. B 38 (1976), no. 1, 1–36. MR 411025

T. W. Anderson and Takamitsu Sawa, Exact and approximate distributions of the maximum likelihood estimator of a slope coefficient, J. Roy. Statist. Soc. Ser. B 44 (1982), no. 1, 52–62. MR 655374

M. Berman, Large sample bias in least squares estimators of a circular arc center and its radius, CVGIP: Image Understanding 45 (1989), 126–128.

A. W. Bowman and A. Azzalini, Applied Smoothing Techniques for Data Analysis: The Kernel Approach with S-Plus Illustrations, Oxford Statistical Science Series, vol. 18, Oxford University Press, 1997.

Raymond J. Carroll, David Ruppert, Leonard A. Stefanski, and Ciprian M. Crainiceanu, Measurement error in nonlinear models, 2nd ed., Monographs on Statistics and Applied Probability, vol. 105, Chapman & Hall/CRC, Boca Raton, FL, 2006. A modern perspective. MR 2243417

Nai Ng Chan, On circular functional relationships, J. Roy. Statist. Soc. Ser. B 27 (1965), 45–56. MR 189163

Chi-Lun Cheng and Alexander Kukush, Non-existence of the first moment of the adjusted least squares estimator in multivariate errors-in-variables model, Metrika 64 (2006), no. 1, 41–46. MR 2242556, DOI https://doi.org/10.1007/s00184-006-0029-z

Chi Lun Cheng and John W. Van Ness, On estimating linear relationships when both variables are subject to errors, J. Roy. Statist. Soc. Ser. B 56 (1994), no. 1, 167–183. MR 1257805

Chi-Lun Cheng and John W. Van Ness, Statistical regression with measurement error, Kendall’s Library of Statistics, vol. 6, Arnold, London; co-published by Oxford University Press, New York, 1999. MR 1719513

N. Chernov, Fitting circles to scattered data: parameter estimates have no moments, Metrika 73 (2011), no. 3, 373–384. MR 2785031, DOI https://doi.org/10.1007/s00184-009-0283-y

Nikolai Chernov, Circular and linear regression, Monographs on Statistics and Applied Probability, vol. 117, CRC Press, Boca Raton, FL, 2011. Fitting circles and lines by least squares. MR 2723019

N. Chernov and C. Lesort, Statistical efficiency of curve fitting algorithms, Comput. Statist. Data Anal. 47 (2004), no. 4, 713–728. MR 2101548, DOI https://doi.org/10.1016/j.csda.2003.11.008

N. Chernov and C. Lesort, Least squares fitting of circles, J. Math. Imaging Vision 23 (2005), no. 3, 239–252. MR 2181705, DOI https://doi.org/10.1007/s10851-005-0482-8

L. J. Gleser, Functional, structural and ultrastructural errors-in-variables models, Proc. Bus. Econ. Statist. Sect. Am. Statist. Ass., 1983, pp. 57–66.

R. Z. Hasminskii and I. A. Ibragimov, On asymptotic efficiency in the presence of an infinite-dimensional nuisance parameter, Probability theory and mathematical statistics (Tbilisi, 1982) Lecture Notes in Math., vol. 1021, Springer, Berlin, 1983, pp. 195–229. MR 735986, DOI https://doi.org/10.1007/BFb0072916

Sabine Van Huffel and Philippe Lemmerling (eds.), Total least squares and errors-in-variables modeling, Kluwer Academic Publishers, Dordrecht, 2002. Analysis, algorithms and applications; Including papers from the 3rd International Workshop on TLS and EIV Modeling held in Leuven, August 27–29, 2001. MR 1951009

J. P. Imhof, Computing the distribution of quadratic forms in normal variables, Biometrika 48 (1961), 419–426. MR 137199, DOI https://doi.org/10.1093/biomet/48.3-4.419

Kenichi Kanatani, Statistical optimization for geometric computation: theory and practice, Machine Intelligence and Pattern Recognition, vol. 18, North-Holland Publishing Co., Amsterdam, 1996. MR 1392697

K. Kanatani, Cramer–Rao lower bounds for curve fitting, Graph. Mod. Image Process. 60 (1998), 93–99.

K. Kanatani, For geometric inference from images, what kind of statistical model is necessary?, Syst. Comp. Japan 35 (2004), 1–9.

K. Kanatani, Optimality of maximum likelihood estimation for geometric fitting and the KCR lower bound, Memoirs Fac. Engin. Okayama Univ. 39 (2005), 63–70.

K. Kanatani, Statistical optimization for geometric fitting: Theoretical accuracy bound and high order error analysis, Int. J. Computer Vision 80 (2008), 167–188.

Alexander Kukush and Erich Otto Maschke, The efficiency of adjusted least squares in the linear functional relationship, J. Multivariate Anal. 87 (2003), no. 2, 261–274. MR 2016938, DOI https://doi.org/10.1016/S0047-259X%2803%2900048-4

A. M. Mathai and Serge B. Provost, Quadratic forms in random variables, Statistics: Textbooks and Monographs, vol. 126, Marcel Dekker, Inc., New York, 1992. Theory and applications. MR 1192786

Yves Nievergelt, A finite algorithm to fit geometrically all midrange lines, circles, planes, spheres, hyperplanes, and hyperspheres, Numer. Math. 91 (2002), no. 2, 257–303. MR 1900920, DOI https://doi.org/10.1007/s002110100326

J. Robinson, The distribution of a general quadratic form in normal variates, Austral. J. Statist. 7 (1965), 110–114. MR 198584

Prasanna Rangarajan and Kenichi Kanatani, Improved algebraic methods for circle fitting, Electron. J. Stat. 3 (2009), 1075–1082. MR 2557129, DOI https://doi.org/10.1214/09-EJS488

E. Zelniker and V. Clarkson, A statistical analysis of the Delogne–Kåsa method for fitting circles, Digital Signal Proc. 16 (2006), 498–522.