×

Robustness and asymptotics of the projection median. (English) Zbl 1461.62072

Summary: The projection median as introduced by the second author and D. Kirkpatrick [Comput. Geom. 42, No. 5, 364–375 (2009; Zbl 1170.65013)] is a robust multivariate, nonparametric location estimator. It is a weighted average of points in a sample, where each point’s weight is proportional to the fraction of directions in which that point is a univariate median. The projection median has the highest possible asymptotic breakdown and is easily approximated in any dimension. Previous works have established various geometric properties of the projection median. In this paper we examine further robustness and asymptotic properties of the projection median. We derive the influence function of the projection median which leads to bounds on the maximum bias and contamination sensitivity, as well as an exact expression for the gross error sensitivity. We discuss the degree to which the projection median satisfies these properties relative to other popular robust estimators: specifically, the Zuo projection median and the half-space median. A method for computing the robustness quantities for any distribution and dimension is provided. We then show that the projection median is strongly consistent and asymptotically normal. A method for estimating and computing the asymptotic covariance of the projection median is provided. Lastly, we introduce a large sample multivariate test of location, demonstrating the use of the aforementioned properties. We conclude that the projection median performs very well in terms of the aforementioned robustness quantities but this comes at the cost of dependence on the coordinate system as the projection median is not affine equivariant.

MSC:

62H12 Estimation in multivariate analysis
62G05 Nonparametric estimation
62G35 Nonparametric robustness
62E20 Asymptotic distribution theory in statistics
65D18 Numerical aspects of computer graphics, image analysis, and computational geometry

Citations:

Zbl 1170.65013

Software:

UCI-ml
Full Text: DOI

References:

[1] Abramowitz, M.; Stegun, I. A., (Handbook of Mathematical Functions With Formulas, Graphs, and Mathematical Tables. Handbook of Mathematical Functions With Formulas, Graphs, and Mathematical Tables, Dover Books on Advanced Mathematics (1974), Dover Publications, Incorporated: Dover Publications, Incorporated New York) · Zbl 0171.38503
[2] Amado, C.; Bianco, A. M.; Boente, G.; Pires, A. M., Robust bootstrap: An alternative to bootstrapping robust estimators, REVSTAT, 12, 2, 169-197 (2014) · Zbl 1314.62090
[3] Bajaj, C., The algebraic degree of geometric optimization problems, Discrete Comput. Geom., 3, 177-191 (1988) · Zbl 0647.90087
[4] Basu, R.; Bhaswar, B.; Tanmoy, T., The projection median of a set of points in \(R^d\), Discrete Comput. Geom., 47, 329-346 (2012) · Zbl 1246.65037
[5] Chen, Z.; Tyler, D. E., The influence function and maximum bias of Tukey’s median, Ann. Statist., 30, 6, 1737-1759 (2002) · Zbl 1015.62058
[6] Cuevas, A.; Fraiman, R., On depth measures and dual statistics. A methodology for dealing with general data, J. Multivariate Anal., 100, 4, 753-766 (2009) · Zbl 1163.62039
[7] Dang, X.; Serfling, R.; Zhou, W., Influence functions of some depth functions, and application to depth-weighted L-statistics, J. Nonparametr. Stat., 21, 1, 49-66 (2009) · Zbl 1154.62041
[8] Donoho, D. L., Breakdown Properties of Multivariate Location Estimators (1982), Harvard University: Harvard University Boston, (Ph.D. thesis)
[9] Donoho, D.; Gasko, M., Breakdown properties of location estimates based on halfspace depth and projected outlyingness, Ann. Statist., 20, 4, 1803-1827 (1992) · Zbl 0776.62031
[10] Durocher, S., Geometric Facility Location under Continuous Motion. Bounded-Velocity Approximations to the Mobile Euclidean k-Centre and k-Median Problems (2006), University of British Columbia, (Ph.D. thesis)
[11] S. Durocher, D. Kirkpatrick, The projection median of a Set of Points in \(R^2\), in: Proceedings of the Seventeenth Canadian Conference on Computational Geometry, Vol. 17, CCCG 2005, 2005, pp. 47-51.
[12] Durocher, S.; Kirkpatrick, D., The projection median of a set of points, Comput. Geom., 42, 364-375 (2009) · Zbl 1170.65013
[13] Durocher, S.; Leblanc, A.; Skala, M., The projection median as a weighted average, J. Comput. Geom., 8, 1, 78-104 (2017) · Zbl 1476.68284
[14] Eaton, M. L., On the projections of isotropic distributions, Ann. Statist., 9, 2, 391-400 (1981) · Zbl 0463.62016
[15] Fang, K.; Kotz, S.; Ng, K., Symmetric Multivariate and Related Distributions (1990), Chapman and Hall/CRC: Chapman and Hall/CRC New York · Zbl 0699.62048
[16] Ghosh, J. K., A new proof of the Bahadur representation of quantiles and an application, Ann. Math. Stat., 42, 6, 1957-1961 (1971) · Zbl 0235.62006
[17] Hampel, F. R.; Ronchetti, E. M.; Rousseeuw, P. J.; Stahel, W. A., Robust Statistics (1986), John Wiley & Sons, Inc.: John Wiley & Sons, Inc. Hoboken, NJ, USA · Zbl 0593.62027
[18] He, X.; Simpson, D. G., Lower bounds for contamination bias: Globally minimax versus locally linear estimation, Ann. Statist., 21, 1, 314-337 (1993) · Zbl 0770.62023
[19] Huber, P. J.; Ronchetti, E. M., Robust Statistics, Wiley Series in Probability and Statistics, 279-287 (2009), John Wiley & Sons, Inc.: John Wiley & Sons, Inc. Hoboken, NJ, USA · Zbl 1276.62022
[20] Kiefer, J., On Bahadur’s representation of sample quantiles, Ann. Math. Stat., 38, 5, 1323-1342 (1967) · Zbl 0158.37005
[21] Kim, J.; Scott, C. D., Robust kernel density estimation, J. Mach. Learn. Res., 13, Sep, 2529-2565 (2012) · Zbl 1436.62119
[22] Liu, R., On a notion of data depth based on random simplices, Ann. Statist., 18, 1, 405-414 (1990) · Zbl 0701.62063
[23] Liu, X., Approximating the projection depth median of dimensions \(p \geq 3\), Comm. Statist. Simulation Comput., 46, 5, 3756-3768 (2017) · Zbl 1368.62054
[24] Liu, R.; Serfling, R.; Souvaine, D., (Data Depth: Robust Multivariate Analysis, Computational Geometry, and Applications. Data Depth: Robust Multivariate Analysis, Computational Geometry, and Applications, DIMACS Series in Discrete Mathematics and Theoretical Computer Science, vol. 72 (2006), American Mathematical Society: American Mathematical Society Providence) · Zbl 1103.62304
[25] Liu, X. H.; Zuo, Y.; Wang, Q. H., Finite sample breakdown point of Tukey’s halfspace median, Sci. China Math., 60, 5, 861-874 (2017) · Zbl 1368.62131
[26] Lopuhaä, H. P.; Rousseeuw, P. J., Breakdown points of affine equivariant estimators of multivariate location and covariance matrices, Ann. Statist., 19, 1, 229-248 (1991) · Zbl 0733.62058
[27] Muller, M. E., A note on a method for generating points uniformly on \(n\)-dimensional spheres, Commun. ACM, 2, 19-20 (1959) · Zbl 0086.11605
[28] Murphy, P. M.; Aha, D. W., UCI Machine Learning Repository (1992), University of California, Irvine, School of Information and Computer Sciences
[29] Ramsay, K. A., Projection median (2017)
[30] Ramsay, K. A., Projection median r pacakge (2017)
[31] Ramsay, K.; Durocher, S.; Leblanc, A., Integrated rank-weighted depth, J. Multivariate Anal., 173, 51-69 (2019) · Zbl 1422.62191
[32] Serfling, R. J., (Approximation Theorems of Mathematical Statistics. Approximation Theorems of Mathematical Statistics, Wiley Series in Probability and Statistics - Applied Probability and Statistics Section Series (1980), John Wiley and Sons Incorporated) · Zbl 0538.62002
[33] Serfling, R., Equivariance and invariance properties of multivariate quantile and related functions, and the role of standardization, J. Nonparametr. Stat., 22, 7, 915-936 (2010) · Zbl 1203.62103
[34] Stahel, W. A., Breakdown of Covariance EstimatorsTechnical Report 31, 1-16 (1981), Fachgruppe für Statistik, Eïdgenoessische Technische Hochschule: Fachgruppe für Statistik, Eïdgenoessische Technische Hochschule Zürich
[35] Street, W. N.; Wolberg, W. H.; Mangasarian, O. L., Nuclear feature extraction for breast tumor diagnosis, (Proceedings of Society of Photo-Optical Instrumentation Engineers, Biomedical Image Processing and Biomedical Visualization, Vol. 1905 (1993))
[36] J.W. Tukey, Mathematics and the picturing of data, in: Proceedings of the International Congress of Mathematicians, Vol. 2, 1975, pp. 523-531. · Zbl 0347.62002
[37] Vardi, Y.; Zhang, C.-H., The multivariate \(L_1\)-median and associated data depth, Proc. Natl. Acad. Sci., 97, 4, 1423-1426 (2000) · Zbl 1054.62067
[38] Zuo, Y., Projection-based depth functions and associated medians, Ann. Statist., 31, 5, 1460-1490 (2003) · Zbl 1046.62056
[39] Zuo, Y., Influence function and maximum bias of projection depth based estimators, Ann. Statist., 32, 1, 189-218 (2004) · Zbl 1105.62350
[40] Zuo, Y., Robustness of weighted \(L^p\)-depth and \(L^p\)-median, Allgemeines Statistisches Archiv, 88, 2, 215-234 (2004) · Zbl 1294.62116
[41] Zuo, Y.; Serfling, R., General notions of statistical depth function, Ann. Statist., 28, 2, 461-482 (2000) · Zbl 1106.62334
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.