A modified version of the usual M-estimation problem is proposed, and sample median is shown to be a solution of this problem for a wide range of choices of the score function. It exposes certain universality in the robustness of sample median in the univariate case, and this property continues to hold even in multivariate set-ups if we consider the multivariate L1-median. Some interesting facts related to this 'modified M-estimation' are discussed, and the consequences of a similar modification of the traditional maximum likelihood approach are explored.Modified M-estimation score function median unbiasedness multivariate L1-median modified maximum likelihood
The sample mean can have poor efficiency relative to various alternative estimators under arbitraril...
International audienceWe compare 43 location estimators as regards their robustness through a Monte ...
A finite sample performance measure of multivariate location estimators is introduced based on "tail...
A modified version of the usual M-estimation problem is proposed, and sample median is shown to be a...
We propose a location estimator based on a convex linear combination of the sample mean and median. ...
The univariate median is a well-known location estimator, which is √n-consistent, asymptotically Gau...
AbstractThe maximum asymptotic bias of an estimator is a global robustness measure of its performanc...
The bachelor thesis deals with the aspect of robustness of estimates, everything is dis- cussed in d...
Univariate median is a well-known location estimator, which is√ n-consistent, asymptotically Gaussia...
The maximum asymptotic bias of an estimator is a global robustness measure of its performance. The p...
Median in some statistical methods Abstract: This work is focused on utilization of robust propertie...
Constrained M (CM) estimates of multivariate location and scatter [Kent, J. T., Tyler, D. E. (1996)....
Nonparametric procedures use weak assumptions such as continuity of the distribution so that they ar...
It is known that the robustness properties of estimators depend on the choice of a metric in the spa...
In this note we study a multivariate extension of the median obtained by considering the median as t...
The sample mean can have poor efficiency relative to various alternative estimators under arbitraril...
International audienceWe compare 43 location estimators as regards their robustness through a Monte ...
A finite sample performance measure of multivariate location estimators is introduced based on "tail...
A modified version of the usual M-estimation problem is proposed, and sample median is shown to be a...
We propose a location estimator based on a convex linear combination of the sample mean and median. ...
The univariate median is a well-known location estimator, which is √n-consistent, asymptotically Gau...
AbstractThe maximum asymptotic bias of an estimator is a global robustness measure of its performanc...
The bachelor thesis deals with the aspect of robustness of estimates, everything is dis- cussed in d...
Univariate median is a well-known location estimator, which is√ n-consistent, asymptotically Gaussia...
The maximum asymptotic bias of an estimator is a global robustness measure of its performance. The p...
Median in some statistical methods Abstract: This work is focused on utilization of robust propertie...
Constrained M (CM) estimates of multivariate location and scatter [Kent, J. T., Tyler, D. E. (1996)....
Nonparametric procedures use weak assumptions such as continuity of the distribution so that they ar...
It is known that the robustness properties of estimators depend on the choice of a metric in the spa...
In this note we study a multivariate extension of the median obtained by considering the median as t...
The sample mean can have poor efficiency relative to various alternative estimators under arbitraril...
International audienceWe compare 43 location estimators as regards their robustness through a Monte ...
A finite sample performance measure of multivariate location estimators is introduced based on "tail...