Abstract:We prove a central limit theorem for an additive functional of the $d$-dimensional fractional Brownian motion with Hurst index $H\in(\frac{1}{2+d},\frac{1}{d})$, using the method of moments, extending the result by Papanicolaou, Stroock and Varadhan in the case of the standard Brownian motion.
What problem does this paper attempt to address?
The problem that this paper attempts to solve is about the central limit theorem (CLT) of an additive functional of fractional Brownian motion (fBm). Specifically, the author aims to prove that when the Hurst exponent \(H\) of the fractional Brownian motion satisfies \(\frac{1}{d + 2}<H<\frac{1}{d}\), the central limit theorem of a certain specific form of additive functional holds. This result is an extension of the results obtained by Papanicolaou, Stroock and Varadhan in the case of standard Brownian motion.
### Main contributions of the paper
1. **Expansion of the Hurst exponent range**: By using the method of moments, the paper extends the existing central limit theorem from the case where \(H\) satisfies \(\frac{1}{d + 1}<H<\frac{1}{d}\) to the case where \(\frac{1}{d + 2}<H\leq\frac{1}{d + 1}\). This is a significant advancement over previous work, because \(\frac{1}{d + 2}\) is the optimal lower bound, and when \(H\leq\frac{1}{d + 2}\), the constant \(C_{H,d}\) will become infinite.
2. **New proof method**: In order to prove the above - mentioned central limit theorem, the author introduces a new method based on Fourier analysis. This method not only provides more accurate estimates, but also makes it more effective when dealing with high - dimensional integrals.
### Specific problem description
The core problem of the paper is to prove the central limit theorem in the following form:
Let \(B(t)=(B_1(t),\ldots,B_d(t))\) be a \(d\)-dimensional fractional Brownian motion, and its Hurst exponent \(H\in(0,1)\). For a function \(f\) that satisfies certain conditions, consider the additive functional:
\[F_n(t)=n^{\frac{Hd - 1}{2}}\int_0^{nt}f(B(s))ds\]
When \(n\) tends to infinity, the distribution of \(F_n(t)\) converges to a random variable:
\[\sqrt{C_{H,d}}\|f\|_{1/H - d}W(L_t(0))\]
where \(W\) is a standard Brownian motion independent of \(B\), \(L_t(0)\) is the local time of \(B\) at the origin, and the constant \(C_{H,d}\) is given by the following formula:
\[C_{H,d}=\frac{2}{(2\pi)^{d/2}}\int_0^{\infty}w^{-Hd}\left(1-\exp\left(-\frac{1}{2}w^{2H}\right)\right)dw = 2^{1-\frac{1}{2H}}\frac{(1 - Hd)}{\pi^{d/2}\Gamma\left(\frac{Hd + 2H - 1}{2H}\right)}\]
### Conclusion
By introducing new Fourier analysis methods and moment methods, the author has successfully proven that the central limit theorem of additive functionals holds in a wider range of Hurst exponents. This result not only extends the scope of application of existing theories, but also provides new tools and methods for the study of fractional Brownian motion.