success rate/probability distribution and uncertainty

27 Views Asked by At

Apologies if this is a question, which is too basic. Despite many google searches, I could not find an answer. Let us say I have a biased coin and success is defined as head. Is there a way to determine the standard error/confidence interval (e.g. +/- 1.96 * se/standard deviation?) around the observed success rate, depending on the number of observations? So if the coin was tossed only 3 times, the confidence interval would be larger than if I tossed it 100 times. I would think this is somewhat related to the binomial distribution but I am interested in the success rate rather than number of successes. I am familiar with bootstrapping, so resampling from the observations may also be an option? Thanks.