Can You Derive the MGF of Y=-X Using MGF of X Directly?

Click For Summary

Discussion Overview

The discussion revolves around deriving the moment generating function (MGF) of the random variable Y defined as Y = -X, using the MGF of another random variable X, denoted as M_X(t). Participants explore whether it is feasible to derive M_Y(t) directly from M_X(t) without first determining the probability density function (pdf) of Y.

Discussion Character

  • Exploratory, Technical explanation, Debate/contested

Main Points Raised

  • One participant introduces the problem of finding M_Y(t) for Y = -X and expresses a desire to avoid deriving the pdf f(y) first.
  • Another participant provides a general formula for the MGF of a random variable and suggests considering the transformation of the variable by a constant.
  • A different participant claims to have found a method involving a specific function x = log(1-x) and mentions that this approach simplifies the MGF calculation, though it depends on the form of the function of x.
  • One participant questions the relevance of the second post to the initial query, indicating a potential disconnect in the discussion.

Areas of Agreement / Disagreement

The discussion remains unresolved, with multiple competing views on how to approach the derivation of M_Y(t) from M_X(t). Participants have not reached a consensus on the best method or the relevance of certain approaches.

Contextual Notes

Limitations include the dependence on specific forms of functions and the potential complexity of the relationship between the MGFs of X and Y. The discussion does not clarify all mathematical steps involved in the proposed methods.

zli034
Messages
106
Reaction score
0
Say r.v. X, we have pdf f(x) and mgf Mx(t) defined.

Then define Y=-X, y is negative x.

Can we get mgf of Y, i.e. My(t) and how?

I know I can go the way to get pdf f(y) first then My(t). I want to know if Mx(t) is already in my hands, it should be easier to get My(t) other than do f(y) first.

I thought about this whole afternoon. Just don't get it. Please help and thanks in advance.

xoxo
 
Physics news on Phys.org
Remember that for any random variable

[tex] M_x(t) = E[e^{tx}][/tex]

If you want the mgf for cX (any constant times X)

[tex] M_{cX}(t) = E[e^{t(cx)}][/tex]

How can you simplify this expression, how does it relate to [tex]M_X(t)[/tex],
and how do both observations relate to your problem?
 
I have found one way.

I have here x=log(1-x) and f(x) also known.

So we can see if I put x=log(1-x) into Mcx(t), it can be simplified easily. Here is my solution for this problem and it depends on the form of function of x.
 
How does your second post relate to your first question?
 

Similar threads

  • · Replies 1 ·
Replies
1
Views
3K
  • · Replies 5 ·
Replies
5
Views
3K
  • · Replies 4 ·
Replies
4
Views
2K
  • · Replies 4 ·
Replies
4
Views
3K
  • · Replies 6 ·
Replies
6
Views
2K
  • · Replies 1 ·
Replies
1
Views
2K
  • · Replies 3 ·
Replies
3
Views
6K
  • · Replies 36 ·
2
Replies
36
Views
5K
  • · Replies 3 ·
Replies
3
Views
3K
  • · Replies 5 ·
Replies
5
Views
3K