Explain why a gamma random variable with parameters $(t, \lambda)$ has an approximately normal distribution when $t$ is large.

What I have come up with so far is:

Let $X=$ the sum of all $X_i$, where $X_1 = X_1, X_2, \ldots , X_i \sim \mathrm{Exp}(\lambda)$. Then because of the CLT as $t$ gets sufficiently large, it can be approximated as a normal distribution.

Is this sufficient or am I missing a key piece of this proof?


That's pretty much it except that you assumed $t$ is an integer and you used "$i$" rather then "$t$" at one point. One way of dealing with non-integer values of $t$ is to go back to the proof of the CLT that uses characteristic functions and make a minor modification in the argument to accomodate non-integers.

(BTW, one writes $Y\sim\mathrm{Exp}$, with that entire expression in MathJax and with \mathrm{} or the like, not $Y$ ~ $Exp$.)


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.