Thread: Standard deviation of the ratio of two means

Hi
I need to calculate the standard deviation of the ratio between two average values (I have the standard deviation of each average).

Steps:
- calculation of ratio F/L (measures are duplicates) of time course of "Wild-Type" and "Mutated" samples
- calculation of average and SD of each sample
- calculation of ratios
1) Wild type/Mutated
2) Wild type/Wild type


I am interested in a formula for Standard deviation calculation of ratios "Wild type/Mutated" and "Wild type/Wild type"
Is it possible?


I got lost googling....

Thanks a lot in advance!!

Maurizio

It sounds like what you want is a propagation of error calculation (you might put "propagation of error" into your favorite search engine).

From my old analytical chemistry text, propagation of error for products/quotients involves using the relative error. Using s for one standard deviation as our measure of error in a mean (m), relative error is s/m. And the overall error in the product is the square root of the sum of the squares of the relative errors. s(final)=sqrt(sumsq(s(i)/m(i)))

thank you for suggestions!
Actually, I don't have a good Statistics Textbook handy, but I will try to delve into the "propagation of error" topic
cheers
Maurizio