geometric distribution

This note last modified November 17, 2021

---

Specific case of the negative binomial distribution s.t. geom(p) = NB(1, p)

It is the distribution of times you do something with success probability p before you succeed

So geom(1/6) is the distribution of the number of times you roll a die before you get a 6

hypothesis testing

Two sample testing using generalized linear models? Not too sure about it, but might be worth looking into…

Testing one sample vs expected mean… idk how to generalize it to two sided test etc, but it looks good