1: \begin{abstract}

2: Regression spline is a useful tool in nonparametric regression.

3: However, finding the optimal knot locations is a known difficult

4: problem. In this article, we introduce the Non-concave Penalized

5: Regression Spline. This proposal

6: method not only produces smoothing spline with optimal convergence

7: rate, but also can adaptively select optimal knots simultaneously.

8: It is insensitive to the number of origin knots.  The method's performance in a

9: simulation has been studied to compare the other methods. The

10: problem of how to choose smoothing parameters, i.e. penalty

11: parameters in the non-concave regression spline is addressed.

12: \end{abstract}