Given a data set represented by vector y=(y1, y2, …, yn), to obtain a profile likelihood for the power p, we consider a grid of values p0, p1, …, pk.