Finding a marginal probability density (stuck at the integration)

MathNugget

Junior Member
Joined
Feb 1, 2024
Messages
195
Hello.
This is f(x,y)=yxy1ey1(0,)(y)1(0,1)(x)f(x, y) = y x^{y-1}e^{-y}\mathbb{1}_{(0, \infty)}(y)\mathbb{1}_{(0, 1)}(x)
Trying to find the marginal probability density of Y.
so I am supposed to solve:
fY(x,y)=yxy1ey1(0,)(y)1(0,1)(x)dx=yey1(0,)(y)xy11(0,1)(x)dx=yey1(0,)(y)01xy1dx=...f_Y(x, y)= \int_{-\infty}^{\infty}y x^{y-1}e^{-y}\mathbb{1}_{(0, \infty)}(y)\mathbb{1}_{(0, 1)}(x)dx = ye^{-y}\mathbb{1}_{(0, \infty)}(y) \int_{-\infty}^{\infty}x^{y-1} \mathbb{1}_{(0, 1)}(x) dx = ye^{-y}\mathbb{1}_{(0, \infty)}(y) \int_{0}^{1}x^{y-1} dx = ...now I suppose 1(0,)(y)\mathbb{1}_{(0, \infty)}(y) tells me y > 0, so
xy1dx=xyy\int x^{y-1} dx = \frac{x^y}{y}01xy1dx=1y?\int_{0}^{1}x^{y-1} dx = \frac{1}{y}?
Is it fY(x,y)=yey1(0,)(y)1y=ey1(0,)(y)?f_Y(x, y)=ye^{-y}\mathbb{1}_{(0, \infty)}(y) \frac{1}{y}= e^{-y}\mathbb{1}_{(0, \infty)}(y)?
 
Last edited:
1)
Ok, thanks.
I realize now, I could get rid of x in the last part and write fY(y)f_Y (y).

I hope it's not against the forum rules, but I have some further questions (I dont want to spam the forum, but they're only half-related as they're part of the same exercise):
(my choice of words to describe these statistics notions might be wrong, I am trying to double check with chat gpt):
If I want to simulate some random variables from this fYf_Y, I know 2 methods: one is to calculate F from f, and then its inverse, and then I have a formula, second is to find another function, g(y) , such that fY(y)f_Y(y)<cg(y)...

I guess I can use the first method on this, but would the usual formula work here?
FY(y)=xfY(y)dy?F_Y(y) = \int_{-\infty}^{x}f_Y(y)dy?I see fYf_Y is very easy to integrate, so I am tempted to use the first method (if it can be applied)
 
Last edited:
I figured I can leave for questions here, it might be more effective:

2) How do I find the density of X with the condition that Y=y?
Is it: fXY(X=xY=y)=fX,Y(x,y)fY(y)f_{X | Y} (X=x | Y=y) = \frac{f_{X,Y}(x, y)}{f_Y(y)}, and then I just do some simple calculations?

Furthermore, when calculating P(XxY=y)\mathbb{P}(X \leq x | Y =y), do I just xfXY(X=xY=y)dx\int_{-\infty}^{x}f_{X | Y} (X=x | Y=y) dx
 
Top