Moment generating function of a continuous variable

Click For Summary

Homework Help Overview

The discussion revolves around finding the moment generating function of a continuous probability density function defined as f(x) = 0.15e^{-0.15x}. Participants are exploring the integration process involved in calculating the moment generating function and addressing convergence issues.

Discussion Character

  • Exploratory, Mathematical reasoning, Assumption checking

Approaches and Questions Raised

  • Participants discuss the integration of the moment generating function and the evaluation of limits at infinity. There is a focus on ensuring the correct conditions for convergence and the implications of variable limits.

Discussion Status

The discussion has progressed through various attempts to evaluate the integral and clarify misunderstandings about convergence. Some participants have provided guidance on evaluating limits and checking assumptions about the function's behavior at infinity. Multiple interpretations regarding the calculation of variance and second moments are being explored.

Contextual Notes

There are constraints regarding the assumptions made about the variable t, specifically that t must be less than 0.15 for the integral to converge. Participants are also navigating potential discrepancies between their calculations and those presented in the textbook.

exitwound
Messages
291
Reaction score
1

Homework Statement



Find the moment generating of:

[tex]f(x)=.15e^{-.15x}[/tex]

Homework Equations



[tex]M_x(t)= \int_{-\infty}^{\infty}{e^{tx}f(x)dx}[/tex]

The Attempt at a Solution



I get down to the point (if I've done my calculus correctly) and gotten:

[tex]\frac{.15e^{(t-.15)x}}{t-.15} \Bigr| _{0}^{\infty}[/tex]

and I don't know how to evaulate this. Am I right this far? And if I am, how does this equate to the answer in the back of the book:

[tex]\frac{.15}{.15-t}[/tex]
 
Physics news on Phys.org
Remember that x is your variable of integration, not t. Assume t<0.15, otherwise the integral won't converge, and just plug in the limits like you would normally do.
 
That's what I'm having trouble with. How do I evaluate it at infinity?
 
What's the limit of e-kx, k>0, as x goes to infinity?
 
-inf?
 
Instead of guessing, try plotting the function or plugging numbers into your calculator to see if you can see what the limit is.
 
I'm not guessing. I plotted it out, and as x goes to -inf, e goes to inf.

Oh. typo above.
 
I think you're screwing up the signs. You have a function of the form e-kx where k>0 (this is why you need t<0.15). What does it do as x goes to positive infinity, which is the upper limit of the integral?
 
Oh. I see. My head wouldn't release the reading error.

[tex]\frac{-.15e^{(t-.15)\infty}}{t-.15} - \frac{-.15e^{(t-.15)0}}{t-.15}[/tex]

[tex]0-\frac{.15}{t-.15} = \frac{-.15}{t-.15}=\frac{.15}{.15-t}[/tex]

Do I have it yet? I'm way too tired to think clearly.
 
  • #10
Yup, you got it!
 
  • #11
Okay. I have to take the second derivative of that to find the Variance of the pdf.

First derivative = [tex]\frac{.15}{(.15-t)^2}[/tex]

Second derivative = [tex]\frac{.3}{(.15-t)^3}[/tex]

I've even confirmed the second derviative with Wolfram Alpha. However, when I plug 0 into it (to find the second moment, or the variance), I get 88.88. But the book says 44.44. Any idea what I'm doing wrong?
 
  • #12
I think the book is wrong. I multiplied f(x) by x2 and integrated to calculate E(x2) and got the same answer you did.
 
  • #13
Oh, wait. Were you looking for E(x2) or the variance? If it's the latter, you need to subtract off the square of the mean.
 
  • #14
Well, V(X) = E(x2)-[E(X)]2 but it also says that if I take the 2nd derivative of the moment generating function, I get V(X) or [itex]\sigma ^2[/itex]
 
  • #15
That's not correct. The moment generating function only allows you to calculate E(xn) by taking the n-th derivative.
 
  • #16
So if E(x2) is the second derivative, I have to subtract off [E(X)]^2, or 44.44, which makes it 44.44 just like the book says.

I think I understand that now. Thanks.
 

Similar threads

Replies
1
Views
3K
Replies
6
Views
3K
  • · Replies 40 ·
2
Replies
40
Views
5K
  • · Replies 31 ·
2
Replies
31
Views
5K
  • · Replies 14 ·
Replies
14
Views
2K
Replies
2
Views
2K
  • · Replies 3 ·
Replies
3
Views
1K
  • · Replies 3 ·
Replies
3
Views
3K
  • · Replies 11 ·
Replies
11
Views
3K
  • · Replies 6 ·
Replies
6
Views
2K