Y=-X if X ~ Ber(1/4): Solving the Mystery

Click For Summary

Discussion Overview

The discussion revolves around the transformation of a random variable \(X\) that follows a Bernoulli distribution with parameter \(p = 1/4\) into a new variable \(Y = -X\). Participants explore the implications of this transformation, questioning the nature of the resulting distribution and whether it retains the properties of a Bernoulli distribution.

Discussion Character

  • Exploratory
  • Debate/contested
  • Mathematical reasoning

Main Points Raised

  • One participant asserts that if \(Y = -X\) and \(X \sim Ber(1/4)\), the probabilities for \(Y\) cannot be negative, raising a fundamental question about the nature of \(Y\).
  • Another participant proposes that \(Y\) could also follow a Bernoulli distribution, suggesting a distribution of \(Y\) as \(Y \sim \begin{cases} 1 - p, & y = 0\\ p, & y = -1 \end{cases}\), but expresses uncertainty in this conclusion.
  • A later reply challenges the classification of \(Y\) as a Bernoulli distribution, stating that a Bernoulli random variable only takes values 0 or 1, and suggests that the original question might have intended to ask for the distribution of \(Y = 1 - X\).
  • Another participant reiterates the concern about the practical utility of the observations and discusses the transformation of independent variables with exponential distributions, drawing parallels to the current problem.
  • One participant expresses confidence in correcting a previous hypothesis, indicating disagreement with the proposed distribution for \(Y\).

Areas of Agreement / Disagreement

Participants do not reach a consensus on the nature of the distribution of \(Y\). There are competing views regarding whether \(Y\) can be classified as a Bernoulli distribution, and some participants express uncertainty about the implications of the transformation.

Contextual Notes

There are unresolved assumptions regarding the transformation of random variables and the definitions of distributions involved. The discussion includes references to the properties of Bernoulli distributions and the implications of transformations on their characteristics.

Dustinsfl
Messages
2,217
Reaction score
5
If \(Y = -X\) and \(X\sim Ber(1/4)\), then what is Y?

I know that
\[
X\sim
\begin{cases}
1 - p, & x = 0\\
p, & x = 1
\end{cases}
\]
where \(p = 0.25\) in this case. What is the negative of \(X\) though. It doesn't make any sense making the probabilities negative.
 
Physics news on Phys.org
I'm not confident in this answer but I would consider this to also follow a Bernoulli distribution by the following:

$Y\sim
\begin{cases}
1 - p, & y = 0\\
p, & y = -1
\end{cases}$

You should look up random variable transformations and maybe you can find some examples with discrete transforms. The examples that come to mind I've done in the past year have all been for continuous distributions and involve using the CDF.
 
Jameson said:
I'm not confident in this answer but I would consider this to also follow a Bernoulli distribution by the following:

$Y\sim
\begin{cases}
1 - p, & y = 0\\
p, & y = -1
\end{cases}$

You should look up random variable transformations and maybe you can find some examples with discrete transforms. The examples that come to mind I've done in the past year have all been for continuous distributions and involve using the CDF.

That is the correct distribution, but it is not Bernoulli. A RV with a Bernoulli distribution takes only the values 0 or 1.

Given dwsmiths history of accuracy of posting questions I would not be supprised if what he was really asked for was the distribution of $Y=1-X$

.
 
zzephod said:
That is the correct distribution, but it is not Bernoulli. A RV with a Bernoulli distribution takes only the values 0 or 1.

Given dwsmiths history of accuracy of posting questions I would not be supprised if what he was really asked for was the distribution of $Y=1-X$

.

Your observation is not particularly useful from a practical point of view. Let's suppose to have two independent variables X and Y with exponential distribution and we want to find the distribution of the variable Z = X - Y. It is clear that it is necessary to determine the distribution of the variable - Y that will still exponential only instead of y appears -y and its domain will include all the $y \le 0$. Same problem of course if X and Y have Bernoulli distribution ...

Kind regards

$\chi$ $\sigma$
 
zzephod said:
Given dwsmiths history of accuracy of posting questions I would not be supprised if what he was really asked for was the distribution of $Y=1-X$

.

Well, I can tell you are incorrect with your hypothesis.
 

Similar threads

  • · Replies 2 ·
Replies
2
Views
4K
  • · Replies 5 ·
Replies
5
Views
2K
  • · Replies 1 ·
Replies
1
Views
1K
  • · Replies 4 ·
Replies
4
Views
3K
  • · Replies 14 ·
Replies
14
Views
2K
  • · Replies 1 ·
Replies
1
Views
2K
  • · Replies 1 ·
Replies
1
Views
2K
  • · Replies 4 ·
Replies
4
Views
2K
  • · Replies 0 ·
Replies
0
Views
2K