I completed the Mock 1 for 2022 Sep exam. There's a question that asks to suggest a distribution to model claim frequency when risks are not independent (e.g. accumulation event) and one of the answers is Negative Binomial distribution. How does Negative Binomial distribution allow for dependence of claims?
Hi I am not sure if this is in any syllabus but the statistics behind claims contagion ends up with a negative binomial distribution. A good source for it is https://www.casact.org/monograph/cas-monograph-no-2 section 3.5. You probably can take it as proven given it is part of the mock.
This was covered in this thread https://www.acted.co.uk/forums/inde...of-modeling-distribution-for-frequency.17149/