Abstract:We propose a monotone approximation scheme for a class of fully nonlinear PDEs called G-equations. Such equations arise often in the characterization of G-distributed random variables in a sublinear expectation space. The proposed scheme is constructed recursively based on a piecewise constant approximation of the viscosity solution to the G-equation. We establish the convergence of the scheme and determine the convergence rate with an explicit error bound, using the comparison principles for both the scheme and the equation together with a mollification procedure. The first application is obtaining the convergence rate of Peng's robust central limit theorem with an explicit bound of Berry-Esseen type. The second application is an approximation scheme with its convergence rate for the Black-Scholes-Barenblatt equation.
What problem does this paper attempt to address?
The main problem that this paper attempts to solve is to propose a monotone approximation scheme to solve a class of fully nonlinear partial differential equations (PDEs), namely the G - equation, and determine the convergence rate of this scheme. Specifically, the paper focuses on:
1. **Constructing a monotone approximation scheme**: The paper proposes a recursive construction method based on piecewise - constant approximation of viscosity solutions for solving the G - equation. This method is very useful when dealing with the characteristics of G - distributed random variables.
2. **Determining the convergence rate**: Through the comparison principle and smoothing treatment, the paper establishes the convergence of this approximation scheme and gives explicit error bounds. This not only proves the effectiveness of the scheme but also provides a specific estimate of the convergence speed.
3. **Applying to the robust central limit theorem**: As one of the main applications, the paper provides for the first time the convergence speed of Peng's robust central limit theorem and gives explicit bounds of the Berry - Esseen type. This result improves all the convergence speeds obtained in the existing literature under different model assumptions.
4. **Applying to the Black - Scholes - Barenblatt equation**: Another application is to provide a numerical approximation scheme and its convergence speed for solving the Black - Scholes - Barenblatt equation, which is a commonly used option pricing model under volatility uncertainty.
### Specific problem analysis
#### 1. **Convergence speed of the robust central limit theorem**
The paper derives the convergence speed of the robust central limit theorem through the following steps:
- **Defining a sequence of random vectors**: Let \(\{(X_i, Y_i)\}_{i \geq 1}\) be a sequence of \(R^d\times R^d\) - valued random vectors defined on \((\Omega, \mathcal{H}, \hat{E})\) such that \((X_1, Y_1)=(X, Y)\) and \((X_{i + 1}, Y_{i + 1})\) is independent of \(\{(X_1, Y_1),\ldots,(X_i, Y_i)\}\).
- **Defining partial sums**: Let \(S_n=\sum_{i = 1}^n\left(\frac{X_i}{\sqrt{n}}+\frac{Y_i}{n}\right)\).
- **Convergence result**: According to Peng's robust central limit theorem, we have \(\lim_{n\rightarrow\infty}\hat{E}[\varphi(S_n)]=\tilde{E}[\varphi(\xi+\zeta)]\), where \((\xi, \zeta)\) satisfies the G - distribution.
The paper obtains the following convergence speeds through the results of Theorem 1.3:
- **Degenerate case**: There exists a constant \(C\) depending only on \(T, C_\varphi, \beta, M_3^X, M_2^Y\) such that \(\left|\hat{E}[\varphi(S_n)]-\tilde{E}[\varphi(\xi+\zeta)]\right|\leq Cn^{-\beta/6}\).
- **Non - degenerate case**: If the second moment of the random vector \(X\) is non - degenerate, that is, \(\sigma^2 :=-\hat{E}[-|X|^2]>0\), and the initial data \(\varphi\in C_b^1(R^d)\), then there exists a constant \(C\) depending only on \(T, C_\varphi, \alpha, M_{2+\alpha}^X, M_2^Y\) such that \(\left|\hat{E}[\varphi(S_n)]-\tilde{E}[\varphi(\xi+\zeta)]\right|\leq Cn^{-\max\{\alpha/2, 1/6\}}\).
#### 2. **Numerical approximation of the Black - Scholes - Barenblatt equation**
The paper applies the proposed approximation scheme.