Determining the distribution function

Click For Summary

Homework Help Overview

The problem involves determining the distribution function from a given density function defined piecewise. The density function is specified for different intervals, and the original poster is attempting to integrate these to find the cumulative distribution function.

Discussion Character

  • Exploratory, Mathematical reasoning, Assumption checking

Approaches and Questions Raised

  • The original poster attempts to integrate the density function over specified intervals but expresses confusion about the correctness of their results, particularly at the boundaries of the intervals.
  • Some participants question the integration steps taken and whether the original poster included all necessary components in their calculations.
  • There is a suggestion to reconsider the integration limits and the need to combine results from different intervals.

Discussion Status

The discussion is ongoing, with participants providing clarifications and corrections regarding the integration process. There is acknowledgment of mistakes and a collaborative effort to guide the original poster towards a more accurate understanding of the distribution function.

Contextual Notes

Participants note that the original poster may have overlooked the contribution from the first part of the density function when integrating over the second interval. There is also mention of a zero contribution from the interval where the density function is not defined.

shan
Messages
56
Reaction score
0
I've gotten a weird answer after doing the problem but I'm stuck as to where I messed up.

The density function is this:
[tex]f_{X} (x) = \frac{1}{6}x[/tex] for [tex]0<x\leq2[/tex]
[tex]= \frac{1}{3}(2x-3)[/tex] for 2<x<3
and 0 otherwise

And the question is to find the distribution function.

So integrating for the first part from 0 to x:
[tex]\int \frac{1}{6}u du = \frac {1}{6} \frac{x^2}{2} = \frac{x^2}{16}[/tex]
for 0<x<=2

I have a big problem with the second, part. This is what I did (integrating from 2 to x):
[tex]\int \frac{1}{3} (2u-3) du = \frac{1}{3} [u^2-3u] = \frac{1}{3} (x^2-3x - (2^2-6)) = \frac{x^2}{3} - x + \frac{2}{3}[/tex]
for 2<x<3

I know my answer for the second part is wrong as when x=2, the distribution function = 0 and when x=3, the distribution function = 2/3. But the distribution shouldn't be broken up like that at x=2 and supposedly at x=3, it should = 1. So did I forget to do something to the end points or did I not integrate properly?
 
Last edited:
Physics news on Phys.org
The distribution function is given by:

[tex]F_X(x) := \int_{-\infty}^x f_X(x) \, dx[/tex]

So when you computed [itex]\int_2^x f_X(x) \, dx[/itex], you computed the wrong thing, and there's no reason you should have gotten the right answer.

P.S. 6*2 is not 16.
 
Hurkyl said:
P.S. 6*2 is not 16.
whoops sorry, typo :blushing: that first part is
[tex]\int_0^x \frac{1}{6}u du = \frac {1}{6} \frac{x^2}{2} = \frac{x^2}{12}[/tex]

Hurkyl said:
The distribution function is given by:

[tex]F_X(x) := \int_{-\infty}^x f_X(x) \, dx[/tex]

So when you computed [itex]\int_2^x f_X(x) \, dx[/itex], you computed the wrong thing, and there's no reason you should have gotten the right answer.
which shows I don't really understand what I'm doing but now that you mentioned it...

is the second part then given by [tex]\int_0^2 \frac{1}{6}u du + \int_2^x \frac{1}{3} (2u-3) du[/tex]? ie I forgot to add the first part of f(x)?
 
Right. Of course, there should also be a [itex]\int_{-\infty}^0 f_X(u) \, du[/itex] component as well. (But you know it's zero, so I suppose that's why you left it out)
 
Thank you very much for your help :)
 

Similar threads

  • · Replies 10 ·
Replies
10
Views
3K
  • · Replies 5 ·
Replies
5
Views
3K
Replies
4
Views
3K
  • · Replies 4 ·
Replies
4
Views
1K
Replies
6
Views
2K
  • · Replies 40 ·
2
Replies
40
Views
5K
Replies
3
Views
2K
  • · Replies 5 ·
Replies
5
Views
1K
  • · Replies 4 ·
Replies
4
Views
2K
Replies
5
Views
3K