CAS Exam MAS-II Spring 2019, Question 20
Study guide: Exam MAS-II
You are analyzing the claim counts for n individual policyholders. Let Y_i denote the claim count for the i^{th} policyholder. You build your model as follows:
\Pr(Y_i = y_i \mid P = p) = (1-p)^{y_i}p
f(p) = \frac{\Gamma(\alpha + \beta)}{\Gamma(\alpha)\Gamma(\beta)}p^{\alpha-1}(1-p)^{\beta-1}
f(\alpha) = \frac{1}{5}e^{-\alpha/5}
f(\beta) = \frac{1}{8}e^{-\beta/8}
Assuming that p is constrained to be between 0 and 1, the conditional posterior distribution of p \mid (\alpha, \beta, y_1, \ldots, y_n) is proportional to K.
Identify which of the following could be K.
- A)
p^{\alpha-1}(1-p)^{\beta-1} - B)
p^{\alpha}(1-p)^{y_i+\beta-1} - C)
\frac{\beta}{8}\frac{\alpha}{5}p^{\alpha+n-1}(1-p)^{\sum y_i+\beta-1} - D)
p^{4+n}(1-p)^{\sum y_i+7} - E) None of the above could be
K.
Answer this question and see the worked solution