[ArXiv] use of the median

The breakdown point of the mean is asymptotically zero whereas the breakdown point of the median is 1/2. The breakdown point is a measure of the robustness of the estimator and its value reaches up to 1/2. In the presence of outliers, the mean cannot be a good measure of the central location of the data distribution whereas the median is likely to locate the center. Common plug-in estimators like mean and root mean square error may not provide best fits and uncertainties because of this zero breakdown point of the mean. The efficiency of the mean estimator does not guarantee its unbiasedness; therefore, a bit of care is needed prior to plugging in the data into these estimators to get the best fit and uncertainty. There was a preprint from [arXiv] about the use of median last week.

   [physics.data-an:0804.0606] J. Levy, Use of the median in Physics and Astronomy

To make a long story short, this paper summarizes the statistical properties of the median and quantiles. Instead of χ2 goodness of fit for the best fit and uncertainty (weighted averaging in some sense), if a median estimator is adopted to estimate the center and uncertainty, this paper would be useful to grasp the properties of median.

Leave a comment