Mack 2000, Hurlimann, Brosius, Friedland, Clark

Question 1

Benktander Pros

Answer 1

1. Lower MSE
2. Better approx. of Bayesian procedure
3. Better than CL - incorporates expectation
4. Better than BF - more weight to actual data

Question 2

Hurlimann vs Benktander Differences

Answer 2

1. Full triangle vs single year
2. Loss ratios vs link ratios
3. ELR from data vs selected ELR
4. Needs measure of exposure

Question 3

Least Squares Negative Values

Answer 3

Negative intercept: Link ratio instead
Negative coeff: Budgeted instead

Question 4

Pros of Least Squares

Answer 4

1. More flexible than link, budgeted, BF
2. Performs better when data has significant year-to-year fluctuations
3. Credibility weighting option

Question 5

Pros of Linear Approx to Bayesian

Answer 5

1. Simple to compute
2. Easy to explain
3. Less dependent on underlying distributions

Question 6

Functions of Reinsurance

Answer 6

Stability
Capacity
CAT
Capital and Solvency Margin
Technical Expertise

Question 7

Reasons for a Commutation - Cedant

Answer 7

Exit a LOB
Manage reserves for transfer/sale
Avoid credit risk
Better manage claims and expenses

Question 8

Reasons for a Commutation - Reinsurer

Answer 8

Terminate relationship
Protect from cedant insolvency
Avoid disputes over future development

Question 9

Sufficient Data

Answer 9

Includes all info needed
Difficult in reinsurance due to:
-Manuscript nature
-Changes at cedant level can affect mix of business

Question 10

Reliable Data

Answer 10

Complete, consistent, timely
Difficult in reinsurance due to:
-Cedants have different systems and terminology
-Bordereau reporting differs by cedant
-Reporting lags
-Manuscript nature

Question 11

AY Aggregation Pros and Cons

Answer 11

Pros
-Easy to achieve
-Losses estimated sooner
-Industry benchmarks
-Can separate large loss events
Cons
-Not true match

Question 12

UY Aggregation Pros and Cons

Answer 12

Pros
-True match
-Can separate pricing and UW changes
Cons
-Takes longer to develop
-Can’t isolate large loss events

Question 13

Finite Risk Reinsurance

Answer 13

-Multi-year contract
-Incorporates time value of money and investment income
-Often for run-offs

Types:
Loss Portfolio Transfer
Adverse Dev Cover - reimbursed for losses excess a retention, no reserve transfer

Question 14

Process Variance vs Parameter Variance

Answer 14

Process: Uncertainty due to randomness in insurance process

Parameter: Uncertainty in estimate of parameters

Question 15

Advantages of Growth Curves

Answer 15

Smooth
No need for evenly spaced triangle
Only 2 parameters

Question 16

LDF vs Cape Cod Methods

Answer 16

LDF overparameterizes

Cape Cod has lower parameter variance since fewer parameters and incorporates more info (EP)

Question 17

Residual Graph Tests for Growth Curves

Answer 17

Dev age - loss emergence curve
Incr. Loss - variance/mean ratio
CY - CY effects