Nonparametric estimation in fractional SDE (Q2330959)

The article refers to the stochastic differential equation \[ X(t)=X_0+\int_{0}^{t}b(X(s))ds+\sigma B(t) \] where \(B\) is a fractional Brownian motion of Hurst index \(H\in]1/2,1[\), \(b(.)\) is a continuous map on \(\mathbb{R}\) and \(\sigma\in \mathbb{R}^*\). The paper investigates the consistency and rate of convergence of the Nadaraya-Watson estimator of the drift function \(b\), denoted in this paper by \(\widehat{b}_{T,h}(x)\). The stochastic integral with respect to \(X\) in this estimator is taken in the sense of Skorohod. Because of difficulties in computing the Skorohod integral, a second estimator where the stochastic integral is taken pathwise is considered. In the preliminary part, one presents two different methods to define stochastic integrals with respect to the fractional Brownian motion. The one is based on the pathwise properties of the fractional Brownian motion and the other one uses the results on Malliavin calculus. Under some assumptions on the drift function \(b\) and on the kernel in the expression of the estimator, results on consistency and rate of convergence of the estimators are proved.