I love the response screening platform. Is there a way to choose the fit y-by-x model to be non parametric, so i can compare non-normal distributions? is that the Huber/cauchy options?
thanks
