Abstract
The trimmed mean is one of the most common estimators of location for symmetrical distributions, whose effect depends on whether the trim rate matches the proportion of contaminated data. Based on the geometric characteristics of the curve of the trimmed variance function, the authors propose two kinds of adaptive trimmed mean algorithms. The accuracy of the estimators is compared with that of other often-used estimates, such as sample mean, trimmed mean, trimean, and median, by means of simulation method. The results show that the accuracy of the adaptive derivative optimization trimmed mean method is close to the optimum performance in case of medium contamination (the contamination rate is less than 50%). Under high contamination situation (the contamination rate equals 80%), the performance of the estimates is comparable to that of the median and is superior to other counterparts.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
References
J. W. Tukey, A survey of sampling from contaminated distributions, Contributions to Probability and Statistics, ed. by Palo Alto, Stanford University Press, Calif., 1960.
J. W. Tukey and D. H. Mclaughlin, Less vulnerable confidence and significance procedures for location based on a single sample: Trimming/Winsorization I, Sankhyā A, 1963, 25: 331–352.
P. J. Huber, Robust estimation of a location parameter, Ann. Math. Statist., 1964, 35: 73–101.
D. C. Hoaglin, F. Mosteller, and J. W. Tukey, Understanding Robust and Exploratory Data Analysis, John Wiley & Sons, New York, 1983.
R. Y. Liu, On a notion of data depth based on random simplices, Ann. Statist., 1990, 18: 405–414.
Y. Zuo, H. Cui, and X. He, On the Stahel-Donoho estimator and depth-weighted means for multivariate data, Ann. Statist., 2004, 32: 169–190.
Y. Zuo, Multivariate trimmed means based on data depth, Statistical Data Analysis Based on the L 1-Norm and Related Methods, ed. by Y. Dodge, Switzerland, 2002.
T. Wang, Y. Li, and H. Cui, On weighted randomly rimmed means, Journal of Systems Science and Complexity, 2007, 20(1): 47–65.
J. Jin and H. Cui, Discriminant analysis based on statistical depth, Journal of Systems Science and Complexity, 2010, 23(2): 362–371.
J. Jin and H. Cui, Asymptotic distributions in the projection pursuit based canonical correlation analysis, Science in China Series A: Mathematics, 2010, 53(2): 485–498.
M. Wu and Y. Zuo, Trimmed and Winsorized means based on a scaled deviation, Journal of Statistical Planning and Inference, 2009, 139: 350–365.
Author information
Authors and Affiliations
Corresponding author
Additional information
This research was supported in part by the National Basic Research Program of China under Grant No. 2010CB950703, the Natural Science Foundation of China under Grant No. 10901020, and the Fundamental Research Funds for the Central Universities.
This paper was recommended for publication by Editor Guohua ZOU.
Rights and permissions
About this article
Cite this article
Li, S., Li, Y. & Jin, J. Adaptive trimmed mean as a location estimate. J Syst Sci Complex 25, 973–986 (2012). https://doi.org/10.1007/s11424-012-1072-7
Received:
Revised:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11424-012-1072-7