Iterative expectation of continuous and discrete distributions

Join the discussion
Ask a follow-up here, or get your own question answered by working scientists, mathematicians and engineers — people, not an autocomplete.
Real named experts · corrections over time · the nuance an AI answer skips
2 replies · 3K views
cielo
Messages
14
Reaction score
0

Homework Statement


Suppose X ~ uniform (0,1) and the conditional distribution of Y given X = x is binomial (n, p=x), i.e. P(Y=y|X=x) = nCy x[tex]^{y}[/tex] (1-x)[tex]^{n-y}[/tex] for y = 0, 1,..., n.

Homework Equations


FInd E(y) and the distribution of Y.

The Attempt at a Solution


f(x) = [tex]\frac{1}{b-a}[/tex] = [tex]\frac{1}{1-0}[/tex] =1E[Y] = E [E[Y|X=x]
= [tex]\int[/tex] E[Y|X=x] f(x) dx where the integral is from o to 1
= [tex]\int[/tex] [[tex]\Sigma[/tex] y f(y|x)] f(x) dx
= [tex]\int[/tex] [[tex]\Sigma[/tex] y nCy x[tex]^{y}[/tex] (1-x)[tex]^{n-y}[/tex]] f(x) dx

...but I do not know anymore what to do next...please help.
 
Physics news on Phys.org
cielo said:

Homework Statement


Suppose X ~ uniform (0,1) and the conditional distribution of Y given X = x is binomial (n, p=x), i.e. P(Y=y|X=x) = nCy x[tex]^{y}[/tex] (1-x)[tex]^{n-y}[/tex] for y = 0, 1,..., n.


Homework Equations


FInd E(y) and the distribution of Y.


The Attempt at a Solution


f(x) = [tex]\frac{1}{b-a}[/tex] = [tex]\frac{1}{1-0}[/tex] =1


E[Y] = E [E[Y|X=x]
= [tex]\int[/tex] E[Y|X=x] f(x) dx where the integral is from o to 1
= [tex]\int[/tex] [[tex]\Sigma[/tex] y f(y|x)] f(x) dx
= [tex]\int[/tex] [[tex]\Sigma[/tex] y nCy x[tex]^{y}[/tex] (1-x)[tex]^{n-y}[/tex]] f(x) dx

...but I do not know anymore what to do next...please help.

You know the conditional distribution of Y given X. Use that to find E[Y | X]. The answer is a function of X - find its expectation with respect to X to get E[E[Y |X]] = E[Y]
 
Thank you so much for your very good idea. Because of that, I already got the E[Y].

Can you still help me in finding the distribution of Y?

I am confused about this one I made:

P[Y] = [tex]\int^{0}_{1}[/tex] [tex]\left[nCy x^{y} (1-x)^{n-y} dx\right][/tex]

I understand that is a a beta function if we ignore the constant. But can you help me find the final distribution of Y?