# Dividing means and their SD to give confidence limits

#### Ox_biochem

##### New Member
Hi all,

I am trying to solve the following elementary stats question, but am not sure how:

"10 values were taken to calculate the mean of 'A', and 10 separate values were taken to calculate the mean of 'B':

mean A = 73.5 (standard deviation 6.3)
mean B = 12.7 (standard deviation 2.7)

Calculate A/B, as well as giving 95% confidence limits on this value"

Any help would be hugely appreciated!

#### Dason

##### Ambassador to the humans
If that's all you were given then you actually can't create a confidence interval since the propogation of error formula for A/B requires the covariance of the two estimates. If you're making the assumption that the individuals in A and B are independent and that the covariance is 0 then it's possible.

But I would suggest you use the search term "propagation of error" or "delta method" and see if you can find the answer yourself (assuming you're willing to assume independence).