When generating uniformly-distributed samples from a multidimensional distribution, I believe that sampling each dimension independently produces uniformly-distributed samples from the original distribution.

For example, If I want to generate uniformly-distributed samples of 3D points, I can sample each axis from the uniform distribution and then combine the results. Sampling dimensions independently from uniform distributions produces results that are also sampled from a uniform distribution.

Is there a name for this principle?

  • 2
    $\begingroup$ I'd say this is an application of Fubini - or just the definition of product measure $\endgroup$ – Hagen von Eitzen Feb 20 at 3:01

Your Answer

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

Browse other questions tagged or ask your own question.