CAS Exam 8 Fall 2013, Question 4
Study guide: Exam 8
An actuary is helping design a new internet liability product that would use industry as a rating factor. Different business types such as restaurants, auto manufacturers, and dairy farms would fall into different industry groups. The actuary wants to create several industry factors from a combination of insurance and demographic data, and use this to classify business types into industry groups.
Part a (1 point)
Describe two reasons that a generalized linear model might not be appropriate for developing industry factors.
Part b (0.25 points)
Describe a benefit that a principal component method would have over a generalized linear model for determining the industry factors.
Part c (1 point)
Briefly describe the major steps in using a cluster analysis to group the industry factors.
Part d (0.75 points)
Describe two test statistics that could be used to determine the optimal number of groups from the cluster analysis. Identify which statistic would be preferred when variables are correlated.
No current syllabus covers this question; it is kept for the record.
Answer this question and see the worked solution
Concepts tested
- K-Means Clustering
- Generalized Linear Model
- Principal Components Analysis
- High Dimensional Variables