Here's a variation of the coupon collector problem from Ross that I'm stuck on:
There are $n$ types of coupons. Each newly obtained coupon is, independently, type $i$ with probability $p_i, i = 1,...,n$. Find the expected number and the variance of the number of distinct types obtained in a collection of k coupons.
I thought maybe I would start with an indicator:
Let $X$ be the total number of distinct coupons and $X_i$ the probability that $i^{th}$ is a new coupon. Then $\sum_{i=1}^{n} X_i = X$. Each indicator is 1 w.p. $x/k$. I don't think this is correct though.