If generating uniformly-distributed random data,
and throwing away results which appear too cluttered,
what would the resulting data be called?

(The generated data may be assumed to be
"truly" random for the purposes of the question)

I labeled this pseudo-uniformly-distributed random data.

The random distribution is no longer uniformly-distributed
since the more cluttered a result is,
the less probability it has of being used,
which ultimately offsets the uniformity of the distribution.

A colleague stated that this term is incorrect,
but could not provide a more acceptable term.

  • $\begingroup$ What do you mean by "too clustered"? Consider giving a definition or a pseudo code. $\endgroup$ – Andrea Nov 17 '16 at 17:28
  • $\begingroup$ @AndreaDiBiagio : Literally, someone generated a vector of 20 points of uniformly-distributed random values, looked at them, and if they didn't additionally appear (to the eye) as 'relatively' uniformly distributed from each other (nothing to do with the probability curve), the person threw them out and regenerated. $\endgroup$ – kando Nov 17 '16 at 17:41

I don't think there is a name for this, although I am not an expert in probability or stochastic processes.

It is definitely not uniformly distributed, because the fact that you remove clumped outcomes means that successive events are not independent.

"Pseudo-uniformly distributed" sounds right, because it captures both the idea the the outcome looks uniform, and that it is not taken from a uniform distribution.


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.