Universal projection theorems with applications to multifractal analysis and the dimension of every ergodic measure on self-conformal sets simultaneously
arXiv:2412.03529
Abstract
We prove a universal projection theorem, giving conditions on a parametrized family of maps and a collection M of measures on X under which for almost every equality holds for all measures simultaneously (i.e. on a full measure set of 's independent of ). We require family to satisfy a transversality condition and collection M to satisfy a new condition called relative dimension separability. Under the same assumptions, we also prove that if the Assouad dimension of X is smaller than d, then for almost every , projection is nearly bi-Lipschitz (i.e. with pointwise -Hölder inverse for every ) at -a.e. x, for all simultaneously. Our setting encompasses families of orthogonal projections, natural projections corresponding to conformal iterated function systems, and non-autonomous or random IFS. As applications, we provide novel results on the multifractal analysis, giving formula for the Hausdorff dimension of a level set of the local dimension for a typical (w.r.t the translation parameter) self-similar measure on the line, valid for the full range spectrum (including the decreasing part of the spectrum; previous results were covering only the increasing part). Among another applications, we prove that given a parametrized contracting conformal IFS satisfying the transversality condition, for almost every parameter the dimension formula holds for all ergodic shift-invariant measures simultaneously. We also prove that the dimension part of the Marstrand's projection theorem holds simultaneously for the collection of all ergodic measures on a strongly separated self-conformal set and for the collection of all Gibbs measures on a self-conformal set (without any separation).
v3: the exposition in the introduction is revised. More details are added in the results on the multifractal analysis (Sections 3.4 and 8)