A bootstrap resampling analysis of galaxy clustering

Abstract

The application of the bootstrap resampling technique (BRT) to determining standard errors is illustrated by means of the galaxy-galaxy correlation function in the Zwicky catalog (1961-68). The BRT involves generating a large number of pseudo-data sets from a single data set and then examining a property of the original set for all the pseudo-data sets. The technique permits a good estimation of the robustness of any calculation involving a property of the original data set. A sample of 1091 14th magnitude galaxies in one sky region is considered in terms of a Poisson sample and pairwise separation. A total of 180 pseudo-data sets was generated, and angular correlations are computed. The results support a significant degree of entropy in the distribution of galaxies.