Eigenmode Analysis of Radial Velocities

Abstract

Radial velocity surveys are examined in terms of eigenmode analysis within the framework of CDM-like family of models. Rich surveys such as MARK III and SFI, which consist of more than $10^{3}$ radial velocities, are found to have a few tens of modes that are not noise dominated. Poor surveys, which have only a few tens of radial velocities, are noise dominated across the eigenmode spectrum. In particular, the bulk velocity of such surveys has been found to be dominated by the more noisy modes. The MARK III and SFI are well fitted by a tilted flat CDM model found by a maximum likelihood analysis and a $χ^{2}$ statistics . However, a mode-by-mode inspection shows that a substantial fraction of the modes lie outside the $90% $ confidence level. This implies that although globally the CDM-like family of models seems to be consistent with radial velocity surveys, in detail it does not. This might indicate a need for a revised power spectrum or for some non-trivial biasing scheme.