Hi everybody, I currently try to solve a problem that ends up with the question "What is the product of two normal distributions?". Unfortunately, I found nothing useful so far.
Now here is my problem: I have two normally distributed variables X and Y. These variables are not independent. The covariance cov(X,Y) = E(XY) - E(X)*E(Y) is not zero. What I would like to know is: How does the covariance change, if I add a normally distributed variable Z (with Mu_Z = 0, Sigma_Z != 0) to Y? Can I calculate cov(X,Y+Z) based on cov(X,Y) and Sigma_Z ? Since E(Y + Z) = E(Y), all I would need is some clou on how E(X*(Y+Z)) looks like. But unfortunately, I don't know so far how X*(Y+Z) looks like. I am thankful for any help! Stefan . . ================================================================= Instructions for joining and leaving this list, remarks about the problem of INAPPROPRIATE MESSAGES, and archives are available at: . http://jse.stat.ncsu.edu/ . =================================================================
