0

Given 2 independent variables, A and B, each following its own Normal Distribution (different means, stdDevs and correlation = 0).

What can be inferred about the division of A by B (A / B)?

Is the "mean" the division of the means of each variable?

Is it possible to have a standard deviation, or at least define a 95% confidence interval?

Boschi
  • 103
  • 1
    Take a simpler case when $A$ and $B$ are independent standard normal random variables, and read the answers to this question. $A/B$ in the case has a Cauchy distribution for which the mean is undefined. – Dilip Sarwate Apr 12 '17 at 01:37
  • And in non-standard case $E|1/B|=+\infty$ too, so the mean of $A/B$ does not exist. – NCh Apr 12 '17 at 10:55

0 Answers0