CAS Exam 7 Spring 2019, Question 11
Study guide: Exam 7
Given the following information as of December 31, 2018:
| Accident Year | Incurred Losses ($000) | Chain Ladder Ultimate ($000) | Bornhuetter-Ferguson Ultimate ($000) | Bayesian Ultimate ($000) |
|---|---|---|---|---|
| 2013 | 4,400 | 4,400 | 4,400 | 4,400 |
| 2014 | 5,000 | 6,000 | 5,833 | 5,840 |
| 2015 | 2,800 | 4,000 | 4,300 | 4,278 |
| 2016 | 3,000 | 5,400 | 5,222 | 5,301 |
| 2017 | 4,000 | 10,000 | 7,000 | 7,750 |
| 2018 | 750 | 3,000 | 4,500 |
\alpha_{2018} = 12{,}500\beta_{2018} = 2.5\varphi_{2018} = 1.1
Part a (0.5 points)
Compare the variance of the prior distributions for accident years 2016 and 2017.
Part b (0.5 points)
Calculate the Bayesian ultimate loss estimate for accident year 2018.
Part c (0.5 points)
Calculate the accident year 2014 correlated chain ladder log mean ultimate loss given the following parameters:
\alpha_{2013} = 8.5172\alpha_{2014} = 8.5172\beta_6 = 0\rho = 0.2
Part d (0.5 points)
Briefly describe one advantage of each of the following methods:
- Correlated chain ladder method
- Bayesian method
Answer this question and see the worked solution
Concepts tested
- Bayesian
- Chain Ladder Method
- Bornhuetter-Ferguson Method
- Bayesian Bornhuetter-Ferguson Model
- Correlated Chain Ladder Model