Let $x_1, x_2, \dots, x_{N}$ i.i.d. random variables $\sim \mathcal{N}\left(0,\sigma^2_x\right)$. Further, let $z\sim \mathcal{N}\left(0,\sigma^2_z\right)$, $z$ is independent from all $x_i$.

We build random variables $y_i=x_i+\gamma z$. By construction, $y_i \sim \mathcal{N}\left(0, \sigma^2_x +\gamma^2 \sigma^2_z\right)$ -- But, $y_i$ are not independent anymore.

What is the following expectation: $$ \mathbb{E} \sum_i \left| y_i\right|, $$ where $\left| a\right|$ is the absolute value of $a$?


Expectation operator distributes over the sum, so correlation amongst $y_i$ is not important in the calculation: $$\mathbb E\left[\sum_{i=1}^N|y_i|\right]=N\mathbb E[|y_1|]$$

$y_1\sim\mathcal N(0, \sigma_y^2)$ where $\sigma_y^2=\sigma_x^2+\gamma^2\sigma_z^2$. You can calculate the expected value using the mean entry of Folded Normal distribution by letting $\mu=0, \sigma^2=\sigma_y^2$. You can choose to calculate it by yourself as well, since the integral will be trivial.

| 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.