Calculate Probability Error for 100 Exp: Sample Size of 15

[mkbh_10]

I have a sample size of 100 experiments out of which 15 are unfavorable events. I do this measurement for different initial conditions of my experiment. The sample size remains constant but the no. of unfavorable events change in each case.

I calculate the probability of success in each case but I want to know the error in this calculation of probability and plot it as error bars in my graph.

How to proceed ?

 

[mfb]

For a fixed (unknown) probability of failure your observed number will follow a binomial distribution, if all 100 experiments are independent of each other.
If you use your observed events to calculate the uncertainty, the result is not exact but unless you have very small numbers of failed experiments the difference will be small. Feldman Cousins would be better in that case.

 

[WWGD]

Maybe you (OP) are trying to construct a confidence interval for the number of failures? Maybe you can use the CLT for the sampling means of experiments 1 through n?