Modified Median Statistics and Type Ia Supernova Data

P. P. Avelino, C. J. A. P. Martins, P. Pinto

The median statistic has recently been discussed by Gott et al. as a more reliable alternative to the standard χ2 likelihood analysis, in the sense that it requires fewer assumptions about the data and is almost as constraining. We apply this statistic to the combined data set of 92 distant Type Ia supernovae currently available and also to a mock SNAP-class data set. We find that the performances of the modified median and χ2 statistics are comparable, particularly in the latter case. We further extend the work of Gott et al. by modifying the median statistic to account for the number and size of sequences of consecutive points above or below the median. We also comment on how the performance of the statistic depends on the choice of free parameters that one is estimating.

The Astrophysical Journal
Volume 575, Page 989
August 2002

DOI: 10.1086/341353