The truncated EM scheme for multiple-delay SDEs with irregular coefficients and application to stochastic volatility model

This paper focuses on the numerical scheme for multiple-delay stochastic differential equations with partially Hölder continuous drifts and locally Hölder continuous diffusion coefficients. To handle with the superlinear terms in coefficients, the truncated Euler-Maruyama scheme is employed. Under the given conditions, the convergence rates at time $T$ in both ${\mathcal{L}}^1$ and ${\mathcal{L}}^2$ senses are shown by virtue of the Yamada-Watanabe approximation technique. Moreover, the convergence rates over a finite time interval $[0,T]$ are also obtained. Additionally, it should be noted that the convergence rates will not be affected by the number of delay variables. Finally, we perform the numerical experiments on the stochastic volatility model to verify the reliability of the theoretical results.