A Comparing Robust Wilks’ statistics in Multivariate Multiple Linear Regression
DOI:
https://doi.org/10.31642/JoKMC/2018/110206Keywords:
Minimum Covariance Determinant Estimator, Outliers, Robustness, P-Value, Wilk's StatisticAbstract
In multivariate linear regression, the classical Wilks’ statistic is the most used method to test hypotheses, which is extremely responsive to the effect of outliers. Many authors have examined the non-robust test statistic established on normal theories for various cases. In this study, we constructed a robust version of Wilks’ statistic relying on a reweighed minimum covariance determinant estimator, that relies on the weight of the observations, with the weights being determined by using the Hampel weight function and Huber weight function. A comparison of the suggested statistics with the regular Wilks’ statistic has been discussed. Monte Carlo studies are used to evaluate how well test statistics work with different datasets. So, this study looked at how two different test statistics perform under a normal distribution. The first is the classical Wilks’ statistic, and the second is a proposed new one. For both of them, the rate of type I errors and the power of the tests were close to the expected significance levels. In situations where the distribution is contaminated, the proposed statistical method works best. If the data has been corrupted or affected somehow, this approach seems to perform the best out of the options
Downloads
References
Leo Breiman and Jerome H Friedman. Predicting multivariate responses in multiple linear regression. Journal of the Royal Statistical Society Series B: Statistical Methodology, 59(1):3–54, 1997. DOI: https://doi.org/10.1111/1467-9868.00054
James B Davis and Joseph W McKean. Rank-based methods for multivariate linear models. Journal of the American Statistical Association, 88(421):245–251, 1993. DOI: https://doi.org/10.1080/01621459.1993.10594316
Esa Ollila, Hannu Oja, and Visa Koivunen. Estimates of regression coefficients based on lift rank covariance matrix. Journal of the American Statistical Association, 98(461):90–98, 2003. DOI: https://doi.org/10.1198/016214503388619120
S Frosch Møller, J¨urgen von Frese, and Rasmus Bro. Robust methods for multivariate data analysis. Journal of Chemometrics: A Journal of the Chemometrics Society, 19(10):549–563, 2005. DOI: https://doi.org/10.1002/cem.962
Ricardo Antonio Maronna. Robust m-estimators of multivariate location and scatter. The annals of statistics, pages 51–67, 1976. DOI: https://doi.org/10.1214/aos/1176343347
Peter J Rousseeuw. Least median of squares regression. Journal of the American statistical association, 79(388):871–880, 1984. DOI: https://doi.org/10.1080/01621459.1984.10477105
P Laurie Davies. Asymptotic behaviour of s-estimates of multivariate location parameters and dispersion matrices. The Annals of Statistics, pages 1269–1292, 1987. DOI: https://doi.org/10.1214/aos/1176350505
Peter J Rousseeuw and Annick M Leroy. A robust scale estimator based on the shortest half. Statistica Neerlandica, 42(2):103–116, 1988. DOI: https://doi.org/10.1111/j.1467-9574.1988.tb01224.x
Hendrik P Lopuhaa. On the relation between s-estimators and m-estimators of multivariate location and covariance. The Annals of Statistics, pages 1662–1683, 1989. DOI: https://doi.org/10.1214/aos/1176347386
David L Woodruff and David M Rocke. Computable robust estimation of multivariate location and shape in high dimension using compound estimators. Journal of the American Statistical Association, 89(427):888–896, 1994. DOI: https://doi.org/10.1080/01621459.1994.10476821
Xuming He and Wing K Fung. High breakdown estimation for multiple populations with applications to discriminant analysis. Journal of Multivariate Analysis, 72(2):151–162, 2000. DOI: https://doi.org/10.1006/jmva.1999.1857
Mia Hubert and Katrien Van Driessen. Fast and robust discriminant analysis. Computational Statistics & Data Analysis, 45(2):301–320, 2004. DOI: https://doi.org/10.1016/S0167-9473(02)00299-2
Curtis A Parvin. An Introduction to Multivariate Statistical Analysis, 3rd ed. T.W. Anderson. Hoboken, NJ: John Wiley amp; Sons, 2003, 742 pp., 99.95, hardcover. ISBN0 − 471 − 36091 − 0. Clinical Chemistry, 50(5): 981 − −982, 052004. DOI: https://doi.org/10.1373/clinchem.2003.025684
Sons Rencher, John Wiley. Methods of Multivariate Analysis, Second Edition. Brigham Young University, 2002. DOI: https://doi.org/10.1002/0471271357
Norm A Campbell. Robust procedures in multivariate analysis i: Robust covariance estimation. Journal of the Royal Statistical Society Series C: Applied Statistics, 29(3):231–237, 1980. DOI: https://doi.org/10.2307/2346896
KC Salter and RF Fawcett. A robust and powerful rank test of treatment effects in balanced incomplete block designs. Communications in Statistics-Simulation and Computation, 14(4):807– 828, 1985. DOI: https://doi.org/10.1080/03610918508812475
Russell Davidson and James G MacKinnon. Graphical methods for investigating the size and power of hypothesis tests. The Manchester School, 66(1):1–26, 1998. DOI: https://doi.org/10.1111/1467-9957.00086
Selcuk Korkmaz, Din¸cer G¨oks¨ul¨uk, and GOKMEN Zararsiz. Mvn: An r package for assessing ¨ multivariate normality. R JOURNAL, 6(2), 2014. DOI: https://doi.org/10.32614/RJ-2014-031
Kanti V Mardia. Measures of multivariate skewness and kurtosis with applications. Biometrika, 57(3):519–530, 1970. DOI: https://doi.org/10.1093/biomet/57.3.519
J Patrick Royston. An extension of shapiro and wilk’s w test for normality to large samples. Journal of the Royal Statistical Society: Series C (Applied Statistics), 31(2):115–124, 1982. DOI: https://doi.org/10.2307/2347973
Patrick Royston. Remark as r94: A remark on algorithm as 181: The w-test for normality. Journal of the Royal Statistical Society. Series C (Applied Statistics), 44(4):547–551, 1995. DOI: https://doi.org/10.2307/2986146
Downloads
Published
Issue
Section
Categories
License
Copyright (c) 2024 Thamer, Abdullah A. Ameen
This work is licensed under a Creative Commons Attribution 4.0 International License.
which allows users to copy, create extracts, abstracts, and new works from the Article, alter and revise the Article, and make commercial use of the Article (including reuse and/or resale of the Article by commercial entities), provided the user gives appropriate credit (with a link to the formal publication through the relevant DOI), provides a link to the license, indicates if changes were made and the licensor is not represented as endorsing the use made of the work.
How to Cite
