On the $\varepsilon$-Euler-Maruyama scheme for time inhomogeneous jump-driven SDEs

Mireille Bossy,Paul Maurer

2024-01-18

Abstract:We consider a class of general SDEs with a jump integral term driven by a time-inhomogeneous random Poisson measure. We propose a two-parameters Euler-type scheme for this SDE class and prove an optimal rate for the strong convergence with respect to the $L^p(\Omega)$-norm and for the weak convergence. One of the primary issues to address in this context is the approximation of the noise structure when it can no longer be expressed as the increment of random variables. We extend the Asmussen-Rosinski approach to the case of a fully dependent jump coefficient and time-dependent Poisson compensation, handling contribution of jumps smaller than $\varepsilon$ with an appropriate Gaussian substitute and exact simulation for the large jumps contribution. For any $p \geq 2$, under hypotheses required to control the $L^p$-moments of the process, we obtain a strong convergence rate of order $1/p$. Under standard regularity hypotheses on the coefficients, we obtain a weak convergence rate of $1/n+\epsilon^{3-\beta}$, where $\beta$ is the Blumenthal-Getoor index of the underlying Lévy measure. We compare this scheme with the Rubenthaler's approach where the jumps smaller than $\varepsilon$ are neglected, providing strong and weak rates of convergence in that case too. The theoretical rates are confirmed by numerical experiments afterwards. We apply this model class for some anomalous diffusion model related to the dynamics of rigid fibres in turbulence.

Probability

What problem does this paper attempt to address?

The paper mainly addresses the following issues:

1. **Construction of Numerical Simulation Algorithm**: The paper proposes a new ε-Euler-Maruyama scheme to simulate jump-type stochastic differential equations (SDEs) driven by time-inhomogeneous random Poisson measures. This scheme can handle situations with complex noise structures and is applicable to a wide range of stochastic models.

2. **Convergence Rate Analysis**: For the proposed numerical scheme, the paper provides theoretical analysis of both strong and weak convergence rates. Specifically, under appropriate assumptions, for any \( p \geq 2 \), strong convergence rate \( O(n^{-1/p}) \) and weak convergence rate \( O(n^{-1}) \) are obtained.

3. **Handling Small Jumps**: When dealing with small jumps, the Asmussen-Rosiński method is used, where an appropriate Gaussian approximation is employed for the small jump part, and the large jump part is simulated accurately. This method can effectively handle jump processes with infinite activity.

4. **Validation through Numerical Experiments**: The effectiveness of the theoretical convergence rates is validated through numerical experiments, and the method is compared with Rubenthaler's method, demonstrating its superiority.

5. **Application Example**: Finally, the paper applies the model to simulate the anomalous diffusion dynamics of rigid fibers in turbulence, showcasing its application value in real physical problems.

In summary, this paper aims to provide an efficient and accurate numerical simulation method for time-inhomogeneous jump-driven stochastic differential equations, and validates its effectiveness through theoretical analysis and numerical experiments.