Abstract
The dynamical discrete web (DyDW), introduced in the recent work of Howitt and Warren, is a system of coalescing simple symmetric one-dimensional random walks which evolve in an extra continuous dynamical time parameter τ. The evolution is by independent updating of the underlying Bernoulli variables indexed by discrete space-time that define the discrete web at any fixed τ. In this paper, we study the existence of exceptional (random) values of τ where the paths of the web do not behave like usual random walks and the Hausdorff dimension of the set of such exceptional τ. Our results are motivated by those about exceptional times for dynamical percolation in high dimension by Häggstrom, Peres and Steif, and in dimension two by Schramm and Steif. The exceptional behavior of the walks in the DyDW is rather different from the situation for the dynamical random walks of Benjamini, Häggstrom, Peres and Steif. For example, we prove that the walk from the origin S0τ violates the law of the iterated logarithm (LIL) on a set of τ of Hausdorff dimension one. We also discuss how these and other results should extend to the dynamical Brownian web, the natural scaling limit of the DyDW.
Original language | English (US) |
---|---|
Pages (from-to) | 2832-2858 |
Number of pages | 27 |
Journal | Stochastic Processes and their Applications |
Volume | 119 |
Issue number | 9 |
DOIs | |
State | Published - Sep 2009 |
Keywords
- Brownian web
- Coalescing random walks
- Dynamical random walks
- Exceptional times
- Hausdorff dimension
- Law of the iterated logarithm
- Sticky random walks
ASJC Scopus subject areas
- Statistics and Probability
- Modeling and Simulation
- Applied Mathematics