If $X$ and $Y$ are independent $N(0,1)$ random variables, then how can I calculate the following probability: $$\mathbf{P}\left(2<\left|X\right|+\left|Y\right|<3\right)?$$ I calculated the density function of $\left|X\right|$ and $\left|Y\right|$. I tried to write the density functions of there sums, but I couldn't continue where I had to integrate $e^{-t^{2}}$. I had to acknowledge this path I followed leads nowhere to solve this problem. Is there any easier way? Anyway, I got $$f_{\left|X\right|+\left|Y\right|}\left(y\right)=\frac{4}{\pi}e^{-\frac{y^{2}}{2}}e^{\frac{1}{4}y^{2}}\sqrt{2\pi}\frac{1}{\sqrt[4]{2}}\left[\Phi\left(\frac{1}{\sqrt{2}}y\right)-\frac{1}{2}\right] $$ if it means something, but I'm not sure it is a good solution. Any easier way?


Hint: Consider the integral $I=\int_a^b f'(x)f(x)\,dx$. Then by integration by parts with $dv=f'(x)\,dx$ and $u=f(x),$ we have $$ I=f(b)^2-f(a)^2-I $$ so $I=\frac{f(b)^2-f(a)^2}{2}$.

Your problem is very similar; try performing integration by parts. I didn't check your pdf of $|X|+|Y|$ too closely, but it's definitely at least the correct form.

  • $\begingroup$ I think i figured out an easier way...tomorrow i will check $\endgroup$ – Kapes Mate Apr 2 '20 at 22:50
  • $\begingroup$ It didnt work, now I have to integrate exp((-u^2-v^2)/2) $\endgroup$ – Kapes Mate Apr 3 '20 at 7:10

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.