Abstract:This work is devoted to studying asymptotic behaviors for Volterra type McKean-Vlasov stochastic differential equations with small noise. By applying the weak convergence approach, we establish the large and moderate deviation principles. In addition, we obtain the central limit theorem and find the Volterra integral equation satisfied by the limiting process, which involves the Lions derivative of the drift coefficient.
What problem does this paper attempt to address?
This paper attempts to solve the problem of the asymptotic behavior of Volterra - type McKean - Vlasov stochastic integral equations (SDEs) with small noise. Specifically, the author studies the following three aspects:
1. **Large Deviation Principle (LDP)**:
- When choosing \(h(\varepsilon)=\frac{1}{\sqrt{\varepsilon}}\), study the large deviation principle of \(\frac{X^{\varepsilon}_{t,\xi}-X^{0}_{t,\xi}}{\sqrt{\varepsilon}h(\varepsilon)}\) with the speed of \(\varepsilon^{-1}\).
- Formula representation: \[h(\varepsilon)=\frac{1}{\sqrt{\varepsilon}}\]
2. **Central Limit Theorem (CLT)**:
- When choosing \(h(\varepsilon) = 1\), study the central limit theorem of \(\frac{X^{\varepsilon}_{t,\xi}-X^{0}_{t,\xi}}{\sqrt{\varepsilon}}\), that is, it converges to a stochastic process that solves a specific Volterra integral equation, which involves the Lions derivative of the drift coefficient.
- Formula representation: \[h(\varepsilon)=1\]
3. **Moderate Deviation Estimate (MDP)**:
- When choosing \(h(\varepsilon)\) such that \(h(\varepsilon)\to\infty\) and \(\sqrt{\varepsilon}h(\varepsilon)\to0\), fill the gap between CLT and LDP and provide a moderate deviation estimate with the speed of \(h^{2}(\varepsilon)\).
- Formula representation: \[h(\varepsilon)\to\infty,\quad\sqrt{\varepsilon}h(\varepsilon)\to0\]
To complete these tasks, the author uses the weak convergence method and identifies the corresponding controlled Volterra equations. The main difficulty lies in dealing with the singularity of the kernel function. To overcome this problem, they apply the extended Kolmogorov continuity criterion.
In addition, the paper also discusses the case where, in the absence of the Volterra kernel (i.e., \(K_{1}=K_{2}=1\)), the equation degenerates into the classical McKean - Vlasov SDEs (also known as distribution - dependent SDEs or mean - field SDEs). In this case, a large amount of literature has studied the asymptotic behavior under small - noise conditions, but the results on the small - noise asymptotic behavior of Volterra - type McKean - Vlasov SDEs are the first of their kind.