1 paper · 1 filter
Emmanuel Guerre, Pascal Lavergne
We propose new data-driven smooth tests for a parametric regression function. The smoothing parameter is selected through a new criterion that favors a large smoothing parameter un…