Chapter 10 Glossary

Question 1

Anti-selection

Answer 1

The preference of some insurance applicants for policies whose underwriting requirements are less stringent than others. Anti-selection occurs when a more profitable business is attracted away from an insurer by a competitor who has found a way of identifying the more profitable segment and offers more attractive terms

Question 2

Events not in data (ENID)

Answer 2

When reserving, you need to allow for all possible events, not just events which have occurred before.

Question 3

Model uncertainty

Answer 3

When modelling data for the purpose of reserving, the risk that an appropriate model has been used is known as model uncertainty. The quantification of model uncertainty is difficult to assess, but by using alternative models the risk can be minimised and hence the level of uncertainty can be assessed by comparing the outputs of alternative models.

Question 4

Parameter uncertainty

Answer 4

When a model is fitted based upon historic data, certain parameters are selected, for example, development factors and associated tail distributions or average cost assumptions. There is the possiibility that parameters do not accurately reflect the underlying statistical process. This uncertainty is known as parameter uncertainty.

Question 5

Process uncertainty

Answer 5

Process uncertainty is the risk inherent in writing business and settling claims in general insurance. The modelling of number and amount of claims will vary from the true value owing to random variation. Process uncertainty is represented by a probability density function. For example, claims in the coming year may take on a range of values with associated likelihood.