# Thread: distribution of the mean of an exponential distribution (mgf)

1. ## distribution of the mean of an exponential distribution (mgf)

Let $\overline{X}$ be the mean of a random sample from the exponential distribution, Exp( $\theta$). Using the mgf technique determine the distribution of $\overline{X}$.

$M_{\overline{X}}(t) = Ee^{t\overline{X}}$

$M_{\overline{X}}(t) = Ee^{\frac{t}{n}\sum{X_i}}$

$M_{\overline{X}}(t) = e^{\frac{1}{n}}(1-\frac{t}{\theta})^{-n}$ I'm sure this is wrong, but I don't know why.

The book has the answer in the back, saying it should be $Gamma(\alpha = n, \beta = \frac{\theta}{n})$, but I don't know how to get there. MGF problems often seem to give me difficulty.

2. Let $Y=\sum_{k=1}^nX_k$, where each $X_k\sim\Gamma(1,\theta)$.
Then by indep $Y\sim\Gamma(n,\theta)$.
Then let $Z=Y/n$ and do a calc one change of variables.

OR if you wish to use MGFs...
$E(e^{\bar X t}) = E(e^{Y(t/n)})=M_Y(t/n)$.
Now, I don't know how you're writing your exponential/gamma densities.
BUT subtitute t/n for t in a $\Gamma(n,\theta)$ MGF and it's over.