2
$\begingroup$

Suppose that X ∼ U $(−π/2,π/2$) . Find the pdf of Y = tan(X). Make sure to define the support of the density function.

My work so far: $F_Y(y) = P(Y < y) = P (tan(x) < y) = P (x < tan^{-1}(y)) = F_X(tan^{-1}(y)) = \int_{-\pi/2}^{tan^{-1}(y)} 1/\pi = tan^{-1}(y)/\pi + 1/2 -> derivative -> 1/(\pi + \pi y^{2})$

This is the cauchy distribution. Thanks for the help in the comments!

$\endgroup$
  • $\begingroup$ Double check the solution to your integral (keep in mind what to integrate with respect to). Also, you computed the CDF. Try taking a derivative and then I think you'll recognize the pdf (after you fix the solution to your integral) $\endgroup$ – Brenton Mar 15 '16 at 1:46
  • $\begingroup$ Did you mean $\mathcal U(-\pi^2, \pi^2)$ or $\mathcal U(-\pi/2, \pi/2)$ ? $\endgroup$ – Graham Kemp Mar 15 '16 at 2:04
  • $\begingroup$ Oops, I messed up in the title there. Editting it into the question $\endgroup$ – Sam N. Mar 15 '16 at 2:12
2
$\begingroup$

So you have:

$$F_X(tan^{-1}(y)) = \int_{-\pi/2}^{tan^{-1}(y)} \frac{1}{\pi} dx = \frac{1}{\pi}\left(tan^{-1}(y) + \frac{\pi}{2}\right) = \frac{1}{\pi}tan^{-1}(y) + \frac{1}{2}$$.

Take a derivative with respect to $y$ and you'll end up with a very familiar distribution

$\endgroup$

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.