Estimation of a quadratic regression functional using the sinc kernel

We use the sinc kernel to construct an estimator for the integrated squared regression function. Asymptotic normality of the estimator at different rates is established, depending on whether the regression function vanishes or not.