Title: A new method for testing isotropy with Shannon entropy

Abstract: We propose a novel method for testing isotropy of a three-dimensional distribution using Shannon entropy. We test the method on some Monte Carlo simulations of isotropic and anisotropic distributions and find that the method can effectively identify and characterize different types of hemispherical asymmetry inputted in a distribution. We generate anisotropic distributions by introducing pockets of different densities inside homogeneous and isotropic distributions and find that the proposed method can effectively quantify the degree of anisotropy and determine the geometry of the pockets introduced. We also considered spherically symmetric radially inhomogeneous distributions which are anisotropic at all points other than the centre and find that such anisotropy can be easily characterized by our method. We use a semi analytic galaxy catalogue from the Millennium simulation to study the anisotropies induced by the redshift space distortions and find that the method can separate such anisotropies from a general one. The method may be also suitably adapted for any two dimensional maps on the celestial sphere to study the hemispherical asymmetry in other cosmological observations.