I am perfectly aware that an answer to this question could take slightly more than the usual, but this thing is making me nut. I don’t get if I make some sort of mistake in terms of calculation or if the authors use a peculiar notation. This is the paper that I am referring to, in particular page 826-828. My problem is with s'(α2)= α2/(1+α2). It seems to me, this cannot be the first derivative of the proposal that has to be made as a function of the parameter α2. Instead, it is simply the explicit version of s(α2)= α2/(1+α2) [then, no derivative around...] Do I miss something? Thanks for your feedbacks.