English

Selections and their Absolutely Continuous Invariant Measures

Dynamical Systems 2013-09-25 v1

Abstract

Let I=[0,1]I=[0,1] and consider disjoint closed regions G1,....,GnG_{1},....,G_{n} in % I\times I and subintervals I1,......,In,I_{1},......,I_{n}, such that GiG_{i} projects onto Ii.I_{i.} We define the lower and upper maps τ1,\tau_{1}, τ2\tau_{2} by the lower and upper boundaries of Gi,i=1,....,n,G_{i},i=1,....,n, respectively. We assume τ1\tau_{1}, τ2\tau_{2} to be piecewise monotonic and preserving continuous invariant measures μ1\mu_{1} and μ2\mu_{2}, respectively. Let % F^{(1)} and F(2)F^{(2)} be the distribution functions of μ1\mu_{1} and μ2.\mu_{2}. The main results shows that for any convex combination FF of % F^{(1)} and F(2)F^{(2)} we can find a map η\eta with values between the graphs of τ1\tau_{1} and τ2\tau_{2} (that is, a selection) such that FF is the η\eta -invariant distribution function. Examples are presented. We also study the relationship of the dynamics of multi-valued maps to random maps.

Keywords

Cite

@article{arxiv.1309.6009,
  title  = {Selections and their Absolutely Continuous Invariant Measures},
  author = {A. Boyarsky and P. Góra and Zh. Li},
  journal= {arXiv preprint arXiv:1309.6009},
  year   = {2013}
}
R2 v1 2026-06-22T01:32:40.825Z