1: \begin{abstract}

2: It is well known that the minimax rates of convergence of

3: nonparametric density and regression function estimation of a

4: random variable measured with error is much slower than the rate in the error free

5: case. Surprisingly, we show that  if one is willing to impose a

6: relatively mild assumption in

7: requiring that the error-prone variable has a compact support, then

8: the results can be greatly improved. We describe new and

9: constructive methods to take full advantage of the compact support

10: assumption via spline-assisted semiparametric methods.

11: We further prove that the new estimator achieves

12: the usual nonparametric

13: rate in estimating both the density and regression functions as if

14: there were no measurement error. The

15: proof involves linear and bilinear operator theories, semiparametric

16: theory, asymptotic analysis regarding Bsplines, as well as

17: integral equation treatments.

18: The performance of the new methods is demonstrated through

19: several simulations and a data example.

20: \end{abstract}

21: