Derivation of the moment generating function of the geometric distribution - why is this wrong?

$\begingroup$

Let $P(k,X)=p(1-p)^{k-1}$.

When deriving the moment generating function I start off as follows:

$E[e^{kt}X]=\sum\limits_{k=1}^{\infty}e^{kt}p(1-p)^{k-1}$.

How I end up rearranging this is as follows:

$\frac{p}{1-p}\sum\limits_{k=1}^{\infty}e^{kt}(1-p)^k=\frac{p}{1-p}\sum\limits_{k=1}^{\infty}(e^{t}(1-p))^k=\frac{p}{1-p}\frac{1}{1-e^t(1-p)}$

I'm obviously not arriving at the correct answer, but I want to know why this derivation is wrong.

$\endgroup$

1 Answer

$\begingroup$

Formula: Let $|q|<1$ then we have $$(\star) \ \ \sum_{k=1}^{\infty} q^k = \frac{q}{1-q}.$$ We use this fact for the calculations of MGF $$\mathbb{E}[e^{tX}] = \frac{p}{1-p}\sum_{k=1}^{\infty}e^{tk}(1-p)^k=\frac{p}{1-p}\sum_{k=1}^{\infty}\left(e^{t}(1-p)\right)^k $$ now let us apply $(\star)$ to obtain $$\frac{p}{1-p}\sum_{k=1}^{\infty}\left(\underbrace{e^{t}(1-p)}_{=:q}\right)^k = \frac{p}{1-p}\frac{\overbrace{e^t(1-p)}^{q}}{\underbrace{1-e^t(1-p)}_{1-q}} = \frac{pe^t}{1-e^t(1-p)}.$$

It is well-defined for all $t< -\ln(1-p)$.

$\endgroup$ 4

Your Answer

Sign up or log in

Sign up using Google Sign up using Facebook Sign up using Email and Password

Post as a guest

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

You Might Also Like