You wrote > Surely this relation is already known...? There are no longer any isolated unknown relations of this type (involving summation/shift only) in this part of the number universe, because ordinary generating functions can be identically exchanged with all parts of the equation, and all of them are rational. Sincerely, ralf