Given a gamma distribution with shape $\alpha=2$ and rate $\lambda=10$, I was first asked to find an expression for $\Bbb E[X^k] \ \forall \ k \in \Bbb N$. Directly computing this, I got $$\Bbb E[X^k]=\frac{\Gamma(k+\alpha)}{\lambda^k\Gamma(\alpha)}.$$
Now I am asked to find an expression for $\Bbb E[X^{-2}]$, which given my parameters, would result in the above expression including $\Gamma(0)$ which is not defined. How should I go about this calculation, and is simply using the moment generating function the only way to do so? I understand that the MGF can only be used for nonnegative k, so I don't think that will work.
Finding negative 2nd moment of gamma distribution
gamma distributiongamma functionmoment-generating-functionsprobabilityprobability distributions
Related Question
- [Math] Solving moment generating function of $Y$ where distribution of $X$ is given and $Y = \ln X$
- [Math] Conditional Moment Generating Function and Iterated Rule
- Understanding the connection between the chi-square and the gamma distribution
- Variance of inverse gamma distribution
- The MGF of $X_{(1)}$ when $X_1,X_2$ are i.i.d Gamma variables
Best Answer
$EX^{-2}=100\int_0^{\infty} \frac 1 x e^{-10x}\, dx$. Since $e^{-10x} \to 1$ as $x \to 0$ and $\int_0^{1} \frac 1 x \, dx=\infty$ it follows that $EX^{-2}=\infty$.