Transform Probability Distribution P to Uniform Distribution

  • Context: Graduate 
  • Thread starter Thread starter mjpam
  • Start date Start date
  • Tags Tags
    Theorem
Click For Summary

Discussion Overview

The discussion revolves around the possibility of transforming a given probability distribution P into a uniform distribution, focusing on both discrete and continuous cases. Participants explore theoretical aspects and seek citations related to this transformation.

Discussion Character

  • Exploratory, Technical explanation, Debate/contested

Main Points Raised

  • One participant recalls a theorem suggesting that a transformation exists from a probability distribution P to a uniform distribution, but is uncertain about the details.
  • Another participant proposes a link to the concept of inverse transform sampling as a potential reference for the transformation.
  • A later reply notes that while the uniform distribution can be transformed to any other distribution via the quantile function, the reverse transformation requires the original distribution to be continuous.
  • One participant questions whether a transformation exists from a discrete probability distribution P_{D} to a discrete uniform distribution U_{D}, suggesting that an inverse transformation T^{-1} could be well-defined.

Areas of Agreement / Disagreement

Participants express uncertainty regarding the conditions under which the transformation from P to a uniform distribution is valid, particularly distinguishing between discrete and continuous cases. There is no consensus on the specifics of the transformation or its applicability.

Contextual Notes

Participants note that the continuity of the probability distribution may be a critical factor in determining the validity of the transformation, but this remains unresolved.

mjpam
Messages
79
Reaction score
0
I vaguely remember a theorem that says, given a probability distribution P, there exists a transform T from P to the uniform (I think discrete or continuous) distribution.

Is this true?

Can anyone provide me with an online citation?
 
Physics news on Phys.org
The uniform distribution can be transformed to any other distribution via the quantile function (i.e. the inverse CDF) but for the reverse the distribution must be continuous (a continuous interval can be mapped to a point but not vice-versa).
 
Would it be true to say thatm given a discrete probability distribution, [itex]P_{D}[/itex], and the discrete uniform distribution, [itex]U_{D}[/itex] there exists transform
[itex]T : P_{D}[/itex] [itex]\to U_{D}[/itex] and its inverse [itex]T^{-1}[/itex] is well-deifned?
 

Similar threads

  • · Replies 1 ·
Replies
1
Views
3K
  • · Replies 1 ·
Replies
1
Views
1K
  • · Replies 5 ·
Replies
5
Views
3K
  • · Replies 31 ·
2
Replies
31
Views
4K
  • · Replies 1 ·
Replies
1
Views
2K
  • · Replies 2 ·
Replies
2
Views
2K
  • · Replies 4 ·
Replies
4
Views
2K
  • · Replies 5 ·
Replies
5
Views
2K
  • · Replies 1 ·
Replies
1
Views
1K
  • Poll Poll
  • · Replies 4 ·
Replies
4
Views
2K