Playing with the tails make all kind of mistakes possible, but that seems like a criticism that would apply to any attempt to identify a distribution based on s

Playing with the tails make all kind of mistakes possible, but that seems like a criticism that would apply to any attempt to identify a distribution based on sample.