1: \begin{abstract}

2: We construct an on-line estimator with equidistant design for

3: tracking a smooth function from Stone-Ibragimov-Khasminskii class.

4: This estimator has the optimal convergence rate of risk to zero in sample size.

5: The procedure for setting coefficients of the estimator

6: is controlled by a single parameter and has a simple numerical

7: solution. The off-line version of this estimator allows

8: to eliminate a boundary layer. Simulation results are given.

9: \end{abstract}

10: