[1007.4347] Searching for a Cosmological Preferred Axis: Un
Posted: August 12 2010
The authors look at preferred directions in two ways. First, they analyse the Union2 supernova dataset for the axis of maximum asymmetry. Second, they compare various directions determined from different cosmological datasets and determine the probability that they are all so close to each other by accident.
The asymmetry direction is determined by fitting the spatially flat [tex]\Lambda[/tex]CDM model separately to two hemispheres and finding the direction which maximises the difference in [tex]\Omega_{m0}[/tex]. By comparison with simulations, the authors conclude that there is no statistical significance to the asymmetry. The directions determined from the supernova data, the CMB dipole, quadrupole and octopole, galaxy flows, and quasar polarization naively look rather close to each other (see table 1 on page 6). The authors get a probability of 1\% for the axes to be so close to each other by accident. (This goes up to 7\% if one drops the CMB axes - for example, the CMB dipole is not independent of local galaxy flows.)
It is not clear how much the result would change if the directions were to have error bars, but the significance could drop a lot. For example, since the the amplitude of the maximum hemispheric asymmetry of the Union2 dataset is not statistically significant, the direction should not matter at all. (I am not familiar with the direction determined from polarization of quasars - perhaps someone can comment on that?)
Nevertheless, it is interesting to have a quantitative analysis of the coincidence of directions.
The asymmetry direction is determined by fitting the spatially flat [tex]\Lambda[/tex]CDM model separately to two hemispheres and finding the direction which maximises the difference in [tex]\Omega_{m0}[/tex]. By comparison with simulations, the authors conclude that there is no statistical significance to the asymmetry. The directions determined from the supernova data, the CMB dipole, quadrupole and octopole, galaxy flows, and quasar polarization naively look rather close to each other (see table 1 on page 6). The authors get a probability of 1\% for the axes to be so close to each other by accident. (This goes up to 7\% if one drops the CMB axes - for example, the CMB dipole is not independent of local galaxy flows.)
It is not clear how much the result would change if the directions were to have error bars, but the significance could drop a lot. For example, since the the amplitude of the maximum hemispheric asymmetry of the Union2 dataset is not statistically significant, the direction should not matter at all. (I am not familiar with the direction determined from polarization of quasars - perhaps someone can comment on that?)
Nevertheless, it is interesting to have a quantitative analysis of the coincidence of directions.