Abstract
The paper proposes two approaches to increase the efficiency of the pMST location and scatter estimator and of the RDELA location and scatter estimator. One approach is deduced from classical reweighting, commonly employed by established robust location and scatter estimators, and the other one is derived from Chebychev’s inequality. Simulation results suggest that both approaches are applicable to increase the efficiency of both estimators. Thereby the classical reweighting approach is outperformed by the approach based on Chebychev’s inequality. Using the latter, the performance of the pMST and RDELA estimator can be brought up to the level of the reweighted minimum covariance determinant and reweighted S-estimator.
Similar content being viewed by others
References
Becker, C., Liebscher, S., Kirschstein, T.: Multivariate outlier identification based on robust estimators of location and scatter. In: Robustness and Complex Data Structures—Festschrift in Honour of Ursula Gather, pp. 103–115 (2013)
Bennett, M., Willemain, T.: Resistant estimation of multivariate location using minimum spanning trees. J. Stat. Comput. Simul. 69, 19–40 (2001)
Croux, C., Haesbroeck, G.: Influence function and efficiency of the minimum covariance determinant scatter matrix estimator. J. Multivar. Anal. 71, 161–190 (1999)
Delaunay, B.: Sur la sphere vide. Izv. Akad. Nauk SSSR, Otdelenie Matematicheskii i Estestvennyka Nauk 7, 793–800 (1934)
Donoho, D.: Breakdown properties of multivariate location estimators. Ph.D. thesis, Department of Statistics, Harvard University (1982)
Donoho, D., Huber, P.: The notion of breakdown point. A Festschrift for Erich Lehmann 157–184 (1983)
Hubert, M., Debruyne, M.: Minimum covariance determinant. Wiley Interdiscip. Rev. Comput. Stat. 2(1), 36–43 (2010)
Joe, H.: Generating random correlation matrices based on partial correlations. J. Multivar. Anal. 97(10), 2177–2189 (2006)
Jungnickel, D.: Graphs, networks and algorithms, 3rd edn. Springer (2008)
Kaban, A.: Non-parametric detection of meaningless distances in high dimensional data. Stat. Comput. 22, 375–385 (2012)
Kirschstein, T., Liebscher, S., Becker, C.: Robust estimation of location and scatter by pruning the minimum spanning tree. J. Multivar. Anal. 120, 173–184 (2013)
Liebscher, S., Kirschstein, T.: Restlos: robust estimation of location and scatter. R package version 0.1-2. (2013)
Liebscher, S., Kirschstein, T., Becker, C.: Rdela—a delaunay-triangulation-based, location and covariance estimator with high breakdown point. Stat. Comput. 1–12 (2013). doi:10.1007/s11222-012-9337-5
Lopuhaä, H., Rousseeuw, P.: Breakdown points of affine equivariant estimators of multivariate location and covariance matrices. Ann. Stat. 19(1), 229–248 (1991)
Penrose, M.: A strong law for the longest edge of the minimal spanning tree. Ann. Probab. 27(1), 246–260 (1999)
Qiu, W., Joe, H.: Cluster Generation: random cluster generation (with specified degree of separation). R package version 1.3.1. (2013)
R Core Team.: R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna (2013)
Rousseeuw, P.: Multivariate estimation with high breakdown point. Math. Stat. Appl. 8, 283–297 (1985)
Rousseeuw, P., Croux, C., Todorov, V., Ruckstuhl, A., Salibian-Barrera, M., Verbeke, T., Koller, M., and Maechler, M.: Robustbase: basic robust statistics. R package version 0.9-8. (2013)
Rousseeuw, P., Leroy, A.: Robust Regression and Outlier Detection. Wiley, New York (1987)
Rousseeuw, P., van Driessen, K.: A fast algorithm for the minimum covariance determinant estimator. Technometrics 41, 212–223 (1999)
Saw, J.G., Yang, M.C., Mo, T.C.: Chebyshev inequality with estimated mean and variance. Am. Stat. 38(2), 130–132 (1984)
Todorov, V., Filzmoser, P.: An object-oriented framework for robust multivariate analysis. J. Stat. Softw. 32(3), 1–47 (2009)
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Liebscher, S., Kirschstein, T. Efficiency of the pMST and RDELA location and scatter estimators. AStA Adv Stat Anal 99, 63–82 (2015). https://doi.org/10.1007/s10182-014-0231-7
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10182-014-0231-7