I have the sample of 100 from Poisson distribution. I have 1000 of similar hypothesis and the meta problem of testing the global hypothesis

$$\bigcap H_{0i}: E(X_i) = 5 \quad \text{vs.} \quad \bigcap H_{1i}: E(Xi) > 5$$ and need to estimate the probability of the type I error for the Bonferroni and Fisher tests.

I tried to do this in R by counting for Bonferroni $1-exp^{-\alpha}$, but Im not sure if it's a correct approach. Cannot find any clue on how to do it for Fisher. Thanks for help!


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.