A new space-time Markov chain approximation for general diffusions on metric graphs is shown to converge in p-Wasserstein distance at explicit rates governed by a thinness quantifier of the subdivision.
General diffusions on the star graph as time-changed Walsh Brownian motion
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
abstract
We establish the representation of general regular diffusions on star-shaped graphs as time-changed Walsh Brownian motions. These are regular continuous Markov processes described locally by a family generalized second order differential operators defined on every edge and a gluing condition at the junction vertex. This allows us to prove two additional results: (i) A representation of diffusions with sticky gluing conditions as time-changes of diffusions governed by the same differential operators but with non-sticky gluing conditions. (ii) An occupation times formula for such diffusions, analogous to the classical It\^o--McKean formula for one-dimensional diffusions. Additionally, we prove two results of independent interest. First, conditions under which a diffusion on the star graph is Feller and Feller--Dynkin, extending classical results for one-dimensional diffusions. Second, the existence uniqueness of solutions to the Dirichlet problem on the unit disk of the star graph for a general diffusion operator and explicit expressions for its solution.
citation-role summary
background 1
citation-polarity summary
fields
math.PR 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
unclear 1representative citing papers
citing papers explorer
-
General diffusions on metric graphs as limits of time-space Markov Chains
A new space-time Markov chain approximation for general diffusions on metric graphs is shown to converge in p-Wasserstein distance at explicit rates governed by a thinness quantifier of the subdivision.