It’s not explicit. Like I said, the terms are highly dependent in reality, but for intuition you can think of a series of variables Xk for k from 1 to N, where Xk equals 1/k with probability 1/√N. And think of N as pretty large.
So most of the time, the sum of these is dominated by a lot of terms with small contributions. But every now and then, a big one hits and there’s a huge spike.
(I haven’t thought very much about what functions of k and N I’d actually use if I were making a principled model; 1/k and 1/√N are just there for illustrative purposes, such that the sum is expected to have many small terms most of the time and some very large terms occasionally.)
It’s not explicit. Like I said, the terms are highly dependent in reality, but for intuition you can think of a series of variables Xk for k from 1 to N, where Xk equals 1/k with probability 1/√N. And think of N as pretty large.
So most of the time, the sum of these is dominated by a lot of terms with small contributions. But every now and then, a big one hits and there’s a huge spike.
(I haven’t thought very much about what functions of k and N I’d actually use if I were making a principled model; 1/k and 1/√N are just there for illustrative purposes, such that the sum is expected to have many small terms most of the time and some very large terms occasionally.)