0
$\begingroup$

Let $X_1,...,X_n$ be $p$-vectors uniformly distributed in the ball $B_r=\{x:\Vert{x}\Vert_2\le r,r\gt 0\}.$

Can someone explain to me what it means by "uniformly distributed in the ball"?

For example, what does $X_i$'s distribution look like? I am having trouble understanding these high-dimensional stuff... .

$\endgroup$
  • $\begingroup$ In this thread Ray Koopman suggested a method how to generate multidimensional random data with any kurtosis from normal's to uniform (flat). $\endgroup$ – ttnphns Apr 23 '17 at 18:22
2
$\begingroup$

I think this should help:

Starting off with a 2D "ball", i.e. a circle. Points are uniformly distributed within a circle.

http://blogs.sas.com/content/iml/2016/03/30/generate-uniform-2d-ball.html

This can then be extended to a 3D "ball", i.e. a sphere and onto higher dimensions.

http://www.statsblogs.com/2016/04/06/generate-points-uniformly-inside-a-d-dimensional-ball/

(I would have posted this as a comment, but I'm 9 short!)

| cite | improve this answer | |
$\endgroup$
  • $\begingroup$ Can you extend this answer by summarizing what can be found behind the links? That makes the anser more useful. As it is now it is short and could be closed. $\endgroup$ – kjetil b halvorsen Apr 23 '17 at 18:11
  • $\begingroup$ Apologies. I wanted to just share the links as a comment, which I believed would be helpful. But I wasn't able to. $\endgroup$ – Jeremy Voisey Apr 23 '17 at 18:14

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy