Abstract:We construct a Hunt process that can be described as an isotropic $\alpha$-stable Lévy process reflected from the complement of a bounded open Lipschitz set. In fact, we introduce a new analytic method for concatenating Markov processes. It is based on nonlocal Schrödinger perturbations of sub-Markovian transition kernels and the construction of two supermedian functions with different growth rates at infinity. We apply this framework to describe the return distribution and the stationary distribution of the process. To handle the strong Markov property at the reflection time, we introduce a novel ladder process, whose transition semigroup encodes not only the position of the process, but also the number of reflections.
What problem does this paper attempt to address?
The problem that this paper attempts to solve is to construct a new mathematical model to describe and analyze the isotropic α - stable Lévy process with reflection properties. Specifically, the author hopes to construct a Hunt process, which can be described as an isotropic α - stable Lévy process reflected from the complement of a non - empty bounded Lipschitz set D.
### Specific problem description:
1. **Construction of the reflection process**:
- The main objective of the paper is to construct a strong Markov process \( X := (X_t, t \geq 0) \), which does not terminate when it first leaves the set \( D \), but restarts at a point \( y \) within \( D \), and this point is distributed according to the given probability measure \( \mu(z, dy) \).
- The reflection occurs at \( \tau_D := \inf\{t > 0 : Y_t \notin D\} \), that is, the moment when \( Y \) first leaves \( D \).
2. **Construction of the transition semigroup**:
- In order to handle the strong Markov property at the reflection time, the author introduces a new ladder process, whose transition semigroup not only encodes the position of the process but also records the number of reflections.
- Through this method, the author is able to describe the return distribution and the stationary distribution of the reflection process.
3. **Challenges and solutions**:
- Since the involved semigroup \( K \) is not a Feller semigroup and is usually neither symmetric nor bounded on \( L^2(D) \), traditional analytical methods cannot be directly applied.
- The author overcomes these challenges by extending the state space to the ladder space \( \mathcal{D} = \mathbb{N}_0\times D \) and using non - local Schrödinger perturbations and two super - median functions with different asymptotic behaviors at infinity.
4. **Comparison with other processes**:
- The author also discusses the case when the isotropic α - stable process is replaced by a Brownian motion and points out that in this case, the reflection process cannot be realized as a Hunt process because it cannot be represented by right - continuous paths.
### Summary:
The core problem of the paper is to successfully construct a Hunt process of an isotropic α - stable Lévy process with reflection properties by introducing new mathematical tools and techniques and to solve the difficulties that cannot be handled by traditional methods. This provides a new approach for studying complex stochastic processes and is of great significance both theoretically and practically.