Probability & Expectation Value of X + Y

[ParisSpart]

we have variables X,Y with f(m,n)=P(X=m,Y=n) with f(0.1)=0.1 f(1.0)=0.1 f(1.1)=0.344
find the expectation value E(X+Y)


i need help because i don't how to start to solve this , if i begin with the definition of the expected value i can't do anything any ideas?

 

[statdad]

If your function f requires two arguments, as you indicate by f(m,n), how can you calculate f(0.1) = 0.1 and so on? do you mean f(0,1) = 0.1, f(1,0) = 0.1, f(1,1) = 0.344?
What values other than 0 and 1 are possible for X and Y, or are these the only possibilities? Make sure you've provided ALL the information accurately.

 

[ParisSpart]

f(m.n) the m and n takes 0 and 1 values ...

 

[statdad]

I will assume my comment about you meaning f(0,1) instead of f(0.1) was correct.
Think about it this way: if X, Y can each be only 0 or 1 there are only 4 possibilities: (0,0), (1,0), (0,1), (1,1). You have the probabilities assigned to three of the four, so you can make a table with two columns (labeled 0 and 1) and two rows (also labeled 0 and 1). In each cell put the appropriate probability:
In cell (0,1) put .1, in cell (1,0) put .1, in cell (1,1) put 0.344. (Since the four probabilities have to sum to 1 you can find the probability for cell (0,0) yourself.)

Now you have two ways to go.
Method A: To calculate the expectation work out the sum

<br /> E(X+Y) = \sum f(m,n) (m+n)<br />

- it will have as many terms as there are cells in the table.

Method B:
The rows of your table have the values for X, the columns the values for Y, so the edges of the table give the marginal distributions of X and Y. Then

<br /> E(X+Y) = E(X) + E(Y)<br />

can be calculated using the appropriate marginal distributions.

Methods A and B give the same answer (when applied correctly) so you can use whichever you prefer.

Set things up and try the work before posting more questions.

 

[ParisSpart]

i don't understand the commands of the table... the commands...

 

[ParisSpart]

can you write again the table?

 

[Ray Vickson]

can you write again the table?

More important: can YOU write the table? If so, do it. If you cannot, tell us why.

 

[ParisSpart]

if i wanted to find the P(X=0) i will find f(0,0) and if yes , why?