Let W be a random variable.
Given the rth moment, $E(W^{r}) = \frac{r!+6^{r}}{2^{r}} = M_w^{(r)}(0)$, $r \geq 0$,
how does one derive $f(w)$?
I know that
$M_w(t)=E(e^{wt}) = \begin{cases} \int_{-\infty}^{\infty} e^{wt} f(w) dw & \text{ if W is continuous} \\ \sum_{w = -\infty}^{\infty} e^{wt} f(w) dw & \text{ if W is discrete} \end{cases}$.
So, I can obviously compute the rth moments and equate them to the rth moments I get from differentiating the arbitrary mgf, but I don't see how that's helpful...
And how do we know if W is discrete or continuous?
And no using Laplace transforms. :P
Never mind I got it. One has to expand $e^{tx}$.
http://en.wikipedia.org/wiki/Moment-generating_function#Calculation