Marginal pdf, what am I doing wrong?

Click For Summary
The discussion revolves around finding the marginal probability density function (pdf) of X from the joint pdf f(x,y) = 49/8 * e^(-3.5y) for the specified ranges. The participant initially miscalculated the constant and the bounds for the marginal pdf, leading to confusion about the integration limits. Clarification was provided that the integration should only occur over the region where f(x,y) is non-zero, specifically where -y < x < y and 0 < y < ∞. The need to correctly visualize the region and set up the integral for y, not x, was emphasized, suggesting potential splits in the integration based on the sign of x. Understanding the correct limits and the region of integration is crucial for accurately determining the marginal pdf.
Rifscape
Messages
41
Reaction score
0

Homework Statement



f(xy)=49/8*e^(−3.5*y) 0 < y < inf and −y < x < y

0 otherwise

a. Find the marginal probability density function of X, fX(x). Enter a formula in the first box, and a number for the second and the third box corresponding to the range of x. Use * for multiplication, / for division, ^ for power, abs for absolute value and exp for exponential function. For example, 3abs(x-5)exp(-x/2) means 3|x-5|e-x/2. Use inf for ∞ and -inf for -∞.

I'm not sure what I am doing wrong here, I keep getting 7/4 and the bounds are from -inf to inf right?

Homework Equations


The marginal pdf equation for x.

The Attempt at a Solution



I did the 8/49*∫e^(-3.5*y) from 0 to inf and got 7/4 with the bounds of x being from 0 to inf. What exactly am I doing wrong?

Or would the bounds be from x to infinity? Any advice is appreciated.

Thanks for the help
 
Last edited:
Physics news on Phys.org
First of all I think you should check the problem is typed correctly. I don't think what you have given is a density function. Second, draw a picture of the region where ##f(x,y)\ne 0##. You need to know that to know the correct limits for your integrals.
 
LCKurtz said:
First of all I think you should check the problem is typed correctly. I don't think what you have given is a density function. Second, draw a picture of the region where ##f(x,y)\ne 0##. You need to know that to know the correct limits for your integrals.

You're right, it was 49/8 not 8/49 for the constant. I tried drawing the picture, I got that the bounds seem to look like its from -inf to inf. But for the distribution wouldn't I have to get x in terms of y? That's why I thought the bounds would be from -x to x, since x < y and y > -x.
 
Rifscape said:
You're right, it was 49/8 not 8/49 for the constant. I tried drawing the picture, I got that the bounds seem to look like its from -inf to inf. But for the distribution wouldn't I have to get x in terms of y? That's why I thought the bounds would be from -x to x, since x < y and y > -x.
Your original statement of the problem had limits: 0 < y < inf and −y < x < y. That is the region where your ##f(x,y)\ne 0##. Now, the definition of the marginal density is$$
f_X(x) = \int_{-\infty}^\infty f(x,y)~dy$$You only have to integrate over the region where ##f(x,y)\ne 0##. For each ##x## you integrate in the ##y## direction. When ##x<0## where does ##y## go? When ##x>0## where does ##y## go? You need to look at your region. So let's see how you set up that integral.
 
LCKurtz said:
Your original statement of the problem had limits: 0 < y < inf and −y < x < y. That is the region where your ##f(x,y)\ne 0##. Now, the definition of the marginal density is$$
f_X(x) = \int_{-\infty}^\infty f(x,y)~dy$$You only have to integrate over the region where ##f(x,y)\ne 0##. For each ##x## you integrate in the ##y## direction. When ##x<0## where does ##y## go? When ##x>0## where does ##y## go? You need to look at your region. So let's see how you set up that integral.

Would I need to split it up and integrate x from -inf to 0, and then integrate x from 0 to inf?
 
LCKurtz said:
When ##x<0## where does ##y## go? When ##x>0## where does ##y## go? You need to look at your region. So let's see how you set up that integral.

Rifscape said:
Would I need to split it up and integrate x from -inf to 0, and then integrate x from 0 to inf?

You aren't integrating ##x##. It it is a ##y## integral. You could start by answering the two questions I asked you above. It would also help if you would describe the region in words so I know you have the correct region in the first place.
 
First, I tried to show that ##f_n## converges uniformly on ##[0,2\pi]##, which is true since ##f_n \rightarrow 0## for ##n \rightarrow \infty## and ##\sigma_n=\mathrm{sup}\left| \frac{\sin\left(\frac{n^2}{n+\frac 15}x\right)}{n^{x^2-3x+3}} \right| \leq \frac{1}{|n^{x^2-3x+3}|} \leq \frac{1}{n^{\frac 34}}\rightarrow 0##. I can't use neither Leibnitz's test nor Abel's test. For Dirichlet's test I would need to show, that ##\sin\left(\frac{n^2}{n+\frac 15}x \right)## has partialy bounded sums...