How Do You Calculate E(XY) for Dependent Variables with Given Observations?

  • Context: Undergrad 
  • Thread starter Thread starter James1990
  • Start date Start date
  • Tags Tags
    Expectation
Click For Summary

Discussion Overview

The discussion revolves around calculating the expected value E(XY) for two dependent random variables X and Y, given a specific number of observations and the sum of their products. The scope includes theoretical reasoning and mathematical approaches to expectation in the context of dependent variables.

Discussion Character

  • Technical explanation, Debate/contested, Mathematical reasoning

Main Points Raised

  • One participant seeks assistance in calculating E(XY) given that X and Y are dependent random variables, with a specific sum of their products and number of observations.
  • Another participant suggests estimating E(Z) where Z=XY using the formula E(Z) = Sum(XY)/n, which they calculate as 1060.84/21.
  • A third participant states that expectation does not require independence, proposing that E(XY) can be calculated as E(X)*E(Y) or Sum(XY)/n, but notes that this may be misleading due to the dependency of X and Y.
  • This same participant emphasizes that E(XY) may not equal E(X)E(Y) if X and Y are dependent, and points out that the lack of specific information about X and Y makes E(X) and E(Y) irrelevant in this context.

Areas of Agreement / Disagreement

Participants express differing views on the calculation of E(XY) for dependent variables, with some suggesting methods that assume independence while others caution against this assumption. The discussion remains unresolved regarding the best approach to take.

Contextual Notes

There are limitations regarding the assumptions made about the distributions of X and Y, as well as the implications of their dependency on the calculation of expectation.

James1990
Messages
2
Reaction score
0
Hi everyone,

I was searching an answer for E(XY), where X and Y are two dependent random variables, number of observations n=21 and Sum(x*y)= 1060.84. Can somebody help me?

It's not mentioned, but I think that each x and y of the distributions have the same probability to occur.
Thank you.
 
Physics news on Phys.org
Since you don't know anything about X and Y individually, you could think in terms of Z=XY. Then the estimate for E(Z) = 1064.84/21.
 
Expectation doesn't require independence so you can just do E(xy)=E(x)*E(y) or in this case, sum(XY)/n
 
randomafk said:
Expectation doesn't require independence so you can just do E(xy)=E(x)*E(y) or in this case, sum(XY)/n
This statement is misleading, E(XY) may not = E(X)E(Y) if they are dependent. However in the problem stated here, nothing in particular is known about X and Y, only the product, so E(X) and E(Y) are irrelevant.
 
Last edited:

Similar threads

  • · Replies 5 ·
Replies
5
Views
3K
  • · Replies 7 ·
Replies
7
Views
2K
  • · Replies 1 ·
Replies
1
Views
2K
  • · Replies 14 ·
Replies
14
Views
2K
  • · Replies 1 ·
Replies
1
Views
3K
  • · Replies 30 ·
2
Replies
30
Views
5K
  • · Replies 43 ·
2
Replies
43
Views
6K
  • · Replies 3 ·
Replies
3
Views
3K
  • · Replies 6 ·
Replies
6
Views
2K
  • · Replies 6 ·
Replies
6
Views
3K