I am having a difficult time using moment generating function properties to prove this: (any direction or key properties will be very helpful)

Let $X_1$, $X_2$, . . . be independent and identically distributed random variables, each with the same distribution as the random variable $X$, and let $N$ be a random variable on the nonnegative integers independent of the $X_i$’s. Let $m_X$(t) denote the moment generating function of $X$ and let $m_N$ (t) denote the moment generating function of $N$. Finally, let $Y$ denote the random sum:

$$Y = \begin{cases} 0 &\text{if }N = 0 \\ X_1 + X_2 + ... + X_n &\text{if } N ≥ 1 \end{cases}$$

By conditioning on $N$, show that the moment generating function of $Y$ is given by:

$$m_Y(t) = m_N(\ln(m_X(t)))$$


The mgf of $Y$ conditional on $N=n$ is $$ M_{Y|N=n}(t)=M_X(t)^n, $$ since $Y$ is a sum of independent random variables each with mgf $M_X(t)$. Using the law of total expectation and the definition of the mgf, the mgf of the unconditional distribution of $Y$ is $$ M_Y(t) = E e^{tY} = E E(e^{tY}|N)=E M_X(t)^N = E e^{N\ln M_X(t)} = M_N(\ln M_X(t)). $$

| cite | improve this answer | |

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.