1: \begin{abstract}

2: We investigate the problem of regression where one is allowed to abstain from predicting.

3: We refer to this framework as {\it regression with reject option} as an extension of classification with reject option.

4: In this context, we focus on the case where the rejection rate is fixed and derive the optimal rule which relies on thresholding the conditional variance function.

5: We provide a semi-supervised estimation procedure of the optimal rule involving two datasets:

6: a first {\it labeled} dataset is used to estimate both regression function and conditional variance function while

7: a second {\it unlabeled} dataset is exploited to calibrate the desired rejection rate.

8: The resulting predictor with reject option is shown to be almost as good as the optimal predictor with reject option both in terms of risk and rejection rate.

9: We additionally apply our methodology with $k$NN algorithm and establish rates of convergence for the resulting $k$NN predictor under mild conditions.

10: %finite sample bounds on the resulting $k$NN predictor under mild conditions.

11: Finally, a numerical study is performed to illustrate the benefit of using the proposed procedure.\\

12: {\bf Keywords: }Regression; Regression with reject option; $k$NN; Predictor with reject option.

13: \end{abstract}

14: