Theory of Probability and Mathematical Statistics
A comparative study for two newly developed estimators for the slope in functional EIV linear model
A. A. Al-Sharadqah
Download PDF
Abstract: Two estimators were recently developed in [1] for the slope of a line in the functional EIV model. Both are unbiased, up to order sigma^4, where sigma is the error standard deviation. One estimator was constructed as a function of the maximum likelihood estimator (MLE). Therefore, it was called Adjusted MLE (AMLE). The second estimator was constructed in a completely different approach. Although both the estimators are unbiased, up to the order sigma^4, the latter estimator is much more accurate than the AMLE. We study here these two estimators more rigorously, and we show why one estimator outperforms the other one.
Keywords: Simple linear regression, Errors-in-Variables models, small-noise model, maximum likelihood estimator, bias correction, mean squared errors.
Bibliography: 1. A. Al-Sharadqah, A new perspective in functional EIV linear model: Part I, Comm. Statist. Theory Methods, 47 (2017), no. 14, 7039-7062.
2. A. Al-Sharadqah, N. Chernov, Statistical analysis of curve fitting methods in Errors-In-Variables models, Theory Probab. Math. Statist., 84 (2011), 4-17.
3. A. Al-Sharadqah, N. Chernov, Q. Huang, Errors-In-Variables regression and the problem of moments, Brazilian Journal of Probability and Statistics, 84 (2013), 401-415.
4. Y. Amemiya, W. A. Fuller, Estimation for the nonlinear functional relationship, Annals Statist., 16 (1988), 147-160.
5. T. W. Anderson, Estimation of linear functional relationships: Approximate distributions and connections with simultaneous equations in econometrics, J. R. Statist. Soc. B, 38 (1976), 1-36.
6. T. W. Anderson, T. Sawa, Distributions of estimates of coefficients of a single equation in a simultaneous system and their asymptotic expansions, Econometrica, 41 (1973), 683-714.
7. T. W. Anderson, T. Sawa, Exact and approximate distributions of the maximum likelihood estimator of a slope coefficient, J. R. Statist. Soc. B, 44 (1982), 52-62.
8. C.-L. Cheng, J. W. Van Ness, Statistical Regression with Measurement Error, Arnold, London, 1999.
9. C. L. Cheng, A. Kukush, Non-existence of the first moment of the adjusted least squares estimator in multivariate errors-in-variables model, Metrika, 64 (2006), 41-46.
10. N. Chernov, Circular and linear regression: Fitting circles and lines by least squares, CRC Monographs on Statistics & Applied Probability, vol. 117, Chapman & Hall, 2010.
11. N. Chernov, Fitting circles to scattered data: parameter estimates have no moments, Metrika, 73 (2011), 373-384.
12. N. Chernov, C. Lesort, Statistical efficiency of curve fitting algorithms, Comp. Stat. Data Anal., 47 (2004), 713-728.
13. L. J. Gleser, Functional, structural and ultrastructural errors-in-variables models, Proc. Bus. Econ. Statist. Sect. Am. Statist. Ass., 1983, 57-66.
14. S. van Huffel, ed., Total Least Squares and Errors-in-Variables Modeling, Kluwer, Dordrecht, 2002.
15. K. Kanatani, Statistical optimization for geometric computation: theory and practice, Elsevier, Amsterdam, 1996.
16. K. Kanatani, Cramer-Rao lower bounds for curve fitting, Graph. Mod. Image Process., 60 (1998), pp. 93-99.
17. K. Kanatani, For geometric inference from images, what kind of statistical model is necessary, Syst. Comp. Japan, 35 (2004), 1-9.
18. A. Kukush, E.-O. Maschke, The efficiency of adjusted least squares in the linear functional relationship, J. Multivar. Anal., 87 (2003), 261-274.
19. J. R. Magnus, H. Neudecker, The commutation matrix: some properties and applications, Annals Statist., 7 (1979), 381-394.
20. K. M. Wolter, W. A. Fuller, Estimation of Nonlinear Errors-in-Variables Models, Annals Statist., 10 (1982), 539-548.
21. E. Zelniker, V. Clarkson, A statistical analysis of the Delogne-K ?asa method for fitting circles, Digital Signal Proc., 16 (2006), 498-522.
