Kybernetika 51 no. 6, 909-922, 2015

Rank tests in regression model based on minimum distance estimates

Radim NavrátilDOI: 10.14736/kyb-2015-6-0909

Abstract:

In this paper a new rank test in a linear regression model is introduced. The test statistic is based on a certain minimum distance estimator, however, unlike classical rank tests in regression it is not a simple linear rank statistic. Its exact distribution under the null hypothesis is derived, and further, the asymptotic distribution both under the null hypothesis and the local alternative is investigated. It is shown that the proposed test is applicable in measurement error models. Finally, a simulation study is conducted to show a good performance of the test. It has, in some situations, a greater power than the widely used Wilcoxon rank test.

Keywords:

measurement errors, minimum distance estimates, rank tests

Classification:

62J05, 62G10

paper.pdf

References:

  1. R. J. Adcock: Note on the method of least squares. The Analyst 4 (1877), 183-184.   DOI:10.2307/2635777
  2. T. Anderson and D. Darling: Asymptotic theory of certain goodness of fit criteria based on stochastic processes. Ann. Math. Statist. 23 (1952), 193-212.   DOI:10.1214/aoms/1177729437
  3. J. P. Buonaccorsi: Measurement Error Models, Methods and Applications. Chapman and Hall/CRC, Boca Raton 2010.   DOI:10.1201/9781420066586
  4. R. J. Carroll, D. Ruppert, L. A. Stefanski and C. M. Crainiceanu: Measurement Error in Nonlinear Models: A Modern Perspective. Chapman and Hall/CRC, Boca Raton 2006.   DOI:10.1201/9781420010138
  5. C. L. Cheng and J. W. van Ness: Statistical Regression with Measurement Error.    DOI:10.1002/1097-0258(20000815)19:15<2077::aid-sim500>3.0.co;2-7
  6. E. F. Drion: Estimation of the parameters of a straight line and of the variances of the variables, if they are both subject to error. Indagationes Math. 13 (1951), 256-260.   DOI:10.1016/s1385-7258(51)50036-7
  7. L. Feng, C. Zou and Z. Wang: Rank-based inference for the single-index model. Statist. Probab. Lett. 82 (2012), 535-541.   DOI:10.1016/j.spl.2011.11.025
  8. M. Friedman: The use of ranks to avoid the assumption of normality implicit in the analysis of variance. J. Amer. Statist. Assoc. 32 (1937), 675-701.   DOI:10.1080/01621459.1937.10503522
  9. W. A. Fuller: Measurement Error Models. John Wiley and Sons, New York 1987.   DOI:10.1002/jae.3950030407
  10. G. H. Golub and C. F. van Loan: An analysis of the total least squares problem. SIAM J. Numer. Anal. 17 (1980), 883-893.   DOI:10.1137/0717073
  11. J. Hájek, Z. Šidák and P. K. Sen: Theory of Rank Tests. Second Edition. Academic Press, New York 1999.   DOI:10.1016/b978-012642350-1/50020-5
  12. H. Hotteling and M. R. Pabst: Rank correlation and tests of significance involving no assumptions of normality. Ann. Math. Statist. 7 (1936), 29-43.   DOI:10.1214/aoms/1177732543
  13. J. Jurečková, H. L. Koul, R. Navrátil and J. Picek: Behavior of R-estimators under measurement errors. To apper in Bernoulli.   CrossRef
  14. J. Jurečková, J. Picek and A. K. Md. E. Saleh: Rank tests and regression rank score tests in measurement error models. Comput. Statist. Data Anal. 54 (2010), 3108-3120.   DOI:10.1016/j.csda.2009.08.020
  15. J. Jurečková, P. K. Sen and J. Picek: Metodological Tools in Robust and Nonparametric Statistics. Chapman and Hall/CRC Press, Boca Raton, London 2013.   CrossRef
  16. H. L. Koul: Weighted Empirical Processes in Dynamic Nonlinear Models. Springer, New York 2002.   DOI:10.1007/978-1-4613-0055-7
  17. D. V. Lindley: Regression lines and the linear functional relationship. Suppl. J. Roy. Statist. Soc. 9 (1947), 218-244.   DOI:10.2307/2984115
  18. R. Navrátil and A. K. Md. E. Saleh: Rank tests of symmetry and R-estimation of location parameter under measurement errors. Acta Univ. Palacki. Olomuc., Fac. rer. nat., Mathematica 50 (2011), 95-102.   CrossRef
  19. E. J. G. Pitman: Lecture Notes on Nonparametric Statistics. Columbia University, New York 1948.   CrossRef
  20. E. L. Scott: Note on consistent estimates of the linear structural relation between two variables. Anal. Math. Stat. 21 (1950), 284-288.   DOI:10.2307/2984115
  21. N. V. Smirnov: Sur la distribution de $\omega^2$ (criterium de m. r. v. mises). C. R. Akad. Sci. Paris 202 (1936), 449-452.   CrossRef
  22. L. Tolmatz: On the distribution of the square integral of the brownian bridge. The Annals of Probab. 30 (2002), 253-269.   DOI:10.1214/aop/1020107767
  23. L. Tolmatz: Addenda: On the distribution of the square integral of the brownian bridge. The Annals of Probab. 31 (2003), 530-532.   CrossRef
  24. F. Wilcoxon: Individual comparisons by ranking methods. Biometrics 1 (1945), 80-83.   DOI:10.2307/3001968