Couret and Venter

Question 1

Mapping type of claims to variable letter

Answer 1

Fatal: V

PT: W

Major: X

Minor: Y

Question 2

Types of letters, Couret and Venter

Answer 2

Capital letters: computed from sample data

Lowercase: estimates of the population means

Question 3

Multi-dimensional credibility

Answer 3

Question 4

Matrix equation for v_i

Answer 4

Question 5

Expected excess loss formula

Answer 5

Question 6

3 options to test the results, Couret and Venter

Answer 6

1. Hazard group injury type rations (V_h)
2. Raw sample data injury type ratios (V_i)
3. Injury type ratios resulting from credibility procedures (v_i^est)

Question 7

Initial SSE test, Couret and Venter

Answer 7

Question 8

Initial SSE test results, Couret and Venter

Answer 8

Credibility procedure best, but not much of an improvement over hazard group ratios

Attributed to the fact that class data by year is volatile, estimators are designed to fit that data

True improvement of credibility procedure masked

Question 9

Quintiles Test procedure

Answer 9

1. Sort injury type relativities in increasing order
2. Group into quintiles so that size roughly equal to the same number of TT claims
3. Calculate weighted average injury type relativity across all classes within each quantile and within the hazard group
4. Use results from step 3 to calculate SSE for each of the 3 methods

Question 10

SSE calculation

Hazard Group Method

Raw Sample Data Method

Credibility Procedure Method

Answer 10

Credibility procedure yields substantial reduction in SSE

Question 11

Why credibility procedure does not show an improvement for Hazard Group A

Answer 11

Group is already very homogenous, so little value in trying to create subgroups to improve estimation accuracy