A functional non-central limit theorem for jump-diffusions with periodic coefficients driven by stable Lévy-noise (Q2471619): Difference between revisions

Let \(\mathbb{R}^d\)-valued process \(X_t\) be the solution of a stochastic differential equation \(dX_t= b(X_t)\,dt+ \sigma(X_t-)dL_t\), \(X_0= x_0\), where \(b\in\mathbb{R}^d\) and \(\sigma\in GL(\mathbb{R}^d)\) are periodic three times continuously differentiable functions, and let \(L_t\) be a \(\mathbb{R}^d\)-valued \(\alpha\)-stable Lévy process with \(1<\alpha< 2\) and symmetric spectral measure. Consider the process \(X^{(n)}= n^{-1/\alpha}(X_{nt}- nt\Pi(b)- x_0)\). It is proved that under some further compactness conditions \(X^{(n)}\) converges in distribution to an \(\alpha\)-stable Lévy process \(X^*\) with an averaged jump measure. A consequence of the strong stalling \(n^{-1/\alpha}\) is a non-existence of an additional diffusion coefficient related to a solution of the associated Poisson problem and called enhancement of diffusivity. This makes a difference with a diffusion driven by the Brownian motion. The result gives a positive answer to the question posed in the book of \textit{A. Bensoussan}, \textit{J.-L. Lions} and \textit{G. Papanicolaou} [Asymptotic analysis for periodic structures, North-Holland Publ. Company (1978; Zbl 0404.35001)].