I calculated the correlation matrixes between each variable and i found that some are correlated some are not ? what should i do to find out the variance of the sum of all random variables divided by a random variable (costs divided by cycle time which is also a random variable)

Here is the correlation matrix i generated thank you very much

  • $\begingroup$ Edit: i found this but i don't know if it is related or solves my problem ?stat.cmu.edu/~hseltman/files/ratio.pdf $\endgroup$ – latrous Sep 11 at 8:04
  • $\begingroup$ I'm guessing that the sum on top goes from $1$ to $Y$ and that $Y$ is an integer. In that case I would suggest you compute the conditional variance given $Y$ and then apply the usual formulas. $\endgroup$ – P. Quinton Sep 11 at 8:08
  • $\begingroup$ @P.Quinton, thank you for your answer, but why conditional variance ? (the goal is to find the mean cost per time unit, so i am divding the total cost per time and all is random variables as they depend on simulation results using as entry random variables ) $\endgroup$ – latrous Sep 11 at 8:14
  • $\begingroup$ If you have the conditional variance then you can find the variance using this en.wikipedia.org/wiki/… so all you have to know is $\mathrm{Var}[U|Y]$ and $\mathbb E[U|Y]$ where $U$ is your thing. $\endgroup$ – P. Quinton Sep 11 at 9:24

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.