Premium
Theory & Methods: Exact Bounds for the Bias of Trimmed Means
Author(s) -
Danielak Katarzyna,
Rychlik Tomasz
Publication year - 2003
Publication title -
australian and new zealand journal of statistics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.434
H-Index - 41
eISSN - 1467-842X
pISSN - 1369-1473
DOI - 10.1111/1467-842x.00262
Subject(s) - truncated mean , mathematics , outlier , statistics , trimmed estimator , parametric statistics , scale (ratio) , econometrics , population , upper and lower bounds , mean squared error , mathematical analysis , consistent estimator , physics , demography , minimum variance unbiased estimator , quantum mechanics , estimator , sociology
Trimmed means remove outliers but affect the estimation efficiency in uncontaminated samples. This paper provides and analyses sharp upper and lower bounds on the bias in the non–parametric estimation of population means by various trimmed means in different scale units. It also presents sharp evaluations of expected trimmed means by the expectations of properly truncated parent distributions.