1: \begin{abstract}

2: This paper is devoted to the parametric estimation of a shift together

3: with the nonparametric estimation of a regression function in a

4: semiparametric regression model. We implement a very efficient and easy

5: to handle Robbins--Monro procedure. On the one hand, we propose a

6: stochastic algorithm similar to that of Robbins--Monro in order to

7: estimate the shift parameter. A preliminary evaluation of the

8: regression function is not necessary to estimate the shift parameter.

9: On the other hand, we make use of a recursive Nadaraya--Watson

10: estimator for the estimation of the regression function. This kernel

11: estimator takes into account the previous estimation of the shift

12: parameter. We establish the almost sure convergence for both

13: Robbins--Monro and Nadaraya--Watson estimators. The asymptotic

14: normality of our estimates is also provided. Finally, we illustrate our

15: semiparametric estimation procedure on simulated and real data.

16: \end{abstract}