1: \begin{abstract}

2: We consider the problem of predicting a real random variable from a

3: functional explanatory variable. The problem is attacked by mean of

4: nonparametric kernel approach which has been recently adapted to this

5: functional context. We derive theoretical results by giving a deep

6: asymptotic study of the behaviour of the estimate, including mean squared

7: convergence (with rates and precise evaluation of the constant terms) as

8: well as asymptotic distribution. Practical use of these results are relying

9: on the ability to estimate these constants. Some perspectives in this

10: direction are discussed. In particular a functional version of wild

11: bootstrapping ideas is proposed and used both on simulated and real

12: functional datasets.

13: \end{abstract}

14: