I am trying to understand what N represents when calculating confidence intervals under bootstrap?
In this scenario, I have a sample with 188 values, so the sample size, SS, is 188. I am interested in the mean of these 188 values.
I bootstrap from this one sample 100 times so the number of bootstrap samples, NS, is 100. So I end up with 100 bootstrap sample means (I calculate the mean of each sample of size 188)
I want to calculate a 95% confidence interval on the mean of the samples (of the 188 values in the sample).
Using the formula:
bootstrap_CI_upper = mean(1000 means of each bootstrap sample) + 1.96*sd(1000 means of each bootstrap sample)/sqrt(N)
,
what is N here? Is it the number of bootstrap samples (100) or the size of each sample (188)?
I think N here is 188 (the size of each sample) but I want to confirm it.
Thank you! p.s. Since I'm new to bootstrapping, please let me know if you know of a good resource to learn about applying the bootstrapping method.