XMM-Newton Science Analysis System


ompsfprofile (ompsfprofile-1.4.1) [xmmsas_20141104_1833-14.0.0]

Meta Index / Home Page / Algorithm / using the list of

Outlier rejection

The outlier rejection algorithm makes use of a robust estimate of the data spread, $Sn$, computed using the formula

$Sn = cn \times 1.1926 \times LOMED_i HIMED_i \mid x_i-x_j \mid$

given in [Rousseeuw, P.J. and Croux, C. (1993), "Alternatives to the Median Absolute Deviation," Journal of the American Statistical Association, Vol. 88, 1273-1283].

Three iterations of this calculation are performed, in each of which the median of the data and the $Sn$ value are computed. Data with an absolute deviation from the median of more than 2.0 times $Sn$ are rejected.