Suppose we are given a sequence of iid Gaussian random variables $\{x_k \}$ with zero mean and unit variance. We create a new random variable $$X_n = e^{\sum_{n\ge k}x_k}.$$ How does one go about calculating the expectation $\mathbb{E}[X_n]$? I know that by definition that the expectation of a random variable $Y$ is $$\mathbb{E}[Y] = \int Y(\omega)\mathbb{P}(d\omega) $$ however, I am less clear on how to properly set up this integral in the case of the expectation of $X_n$, as it was created from $n$ random variables $x_k$.



  • You should be able to find the mean and variance of the sum $S$ of $n$ independent random variables with standard normal distributions; it too will have a normal distribution.

  • So you should be able to state the probability density function $f(s)$ for $S$

  • You want to find $$\mathbb E\left[e^S\right] = \int \limits_{-\infty}^{\infty}e^s f(s)\, ds$$ which should not be too difficult if you complete the square inside the exponential

  • There may be shortcuts, either using the moment generating function of a normal distribution or the mean of a lognormal distribution

| cite | improve this answer | |
  • $\begingroup$ This is exactly what I was looking for, thank you for the hints and clearing up my thinking. I appreciate it. $\endgroup$ – TheRefrigerator Apr 24 '19 at 21:30

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.