I am trying to better understand this statement and the assumptions made:

If $X_1,X_2,\ldots$ are non-negative independently and identically distributed random variables with $P(X_i>0)>0$, then $\displaystyle P\left(\sum_{i=1}^\infty X_i=\infty\right)=1$.

Could someone provide simple examples showing that the statement would not hold if: 1) the $X_i$ were identically distributed but not independent and 2) the $X_i$ were independent but not identically distributed? Thanks!

  1. Let $X$ be an arbitrary non-negative random variable such that $\mathbb{P}(X>0) \in (0,1)$. Define $$X_i := X \qquad \text{for all} \, \, i \in \mathbb{N}.$$ Then the sequence is identically distributed (but not independent) and $$\mathbb{P} \left( \sum_{i=1}^{\infty} X_i = \infty \right) = \mathbb{P}(X>0) \in (0,1).$$
  2. The sequence $(X_i)_{i \in \mathbb{N}}$ defined by $$X_i = \frac{1}{i^2}$$ is a sequence of independent random variables and $$\mathbb{P} \left( \sum_{i=1}^{\infty} X_i = \infty \right)=0.$$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.