🚀 go-pugleaf

I'm looking primarily for recommendations on decent advanced texts
on probability and noise analysis of random variables. The guys
in my group all have had undergrad training, game theory, 
Bayesian stats, MLE, etc., but we need to do some serious work on the
following setup. So if you have any direct advice on solutions, that
would be appreciated as well.

Basic problem:
Find the variance (as well as mean, and maybe skew) of 

x/y , where x and y are random variables with known distribution function
(generally Gaussian, maybe Poisson).Noise in x and y is uncorrelated.

Extended problem: same as above, except express x and y as

x = xmean + cornoise +uncornoise

y = ymean + cornoise +uncornoise

where *mean is the expectation of the R.V., cornoise is a correlated
noise term ( correlated between x and y ) and uncornoise is an uncorrelated
noise term in each R.V.

Please email all responses, and if any interest shown I'll post a summary.
Thanks again.

  Darwin's Dad (Carl Witthoft)  @ Adaptive Optics Assoc.,
 54 Cambridgepark Dr.,  Cambridge, MA 02140	617-864-0201
  ...!{decvax,linus,ima,ihnp4,wjh12,wanginst}!bbncca!aoa!carl
  " Buffet-Crampon R-13 , VanDoren B-45, and VanDoren Fortes ."