Chapter18-Modelling

Question 1

Framework

Answer 1

1 Definition of a model
2 Balanced model
3 Model vs formula
4 Obtaining a model and decision process (FENCED)
5 Operational issues when designing a model (SCARCER FILES)
6 Factors to consider in setting a time period for a model
7 Deterministic vs stochastic models
8 Dynamic model
9 Steps in a deterministic model
10 Additional steps in a stochastic model
11 Model points
12 Instances where model points are inappropriate
13 Discount rate in a model
14 Assessing statistical risk
15 Using a model to set premiums
16 Model points don’t all need to be profitable
17 Unmarketable premiums reconsideration
18 Factor in setting premiums (besides profitability and marketability)
19 Assessing capital requirements and return on capital
19 Using models for risk management
20 Suitability of stochastic models for modelling guarantees
21 Variability is not completed modelled in a stochastic model
22 Model error and parameter error
23 Allowing for a risk in a model

Question 2

Operational issues when designing a model

Answer 2

Simple but retains key features

Clear results

Adequately documented

Range of implementation methods

Communicable workings and output

Easy to understand

Refineable and developable

Frequency of cashflows (balance accuracy vs practicality)

Independent verification of outputs

Length of run not too long

Expense not too high

Sensible joint behaviour of variable

Question 3

Deterministic vs stochastic models

Answer 3

Deterministic models

 Quicker, cheaper and easier to design, build and run

 Clearer what scenarios have been tested

 Results are easier to explain to a non-technical audience

Stochastic models

 Allow naturally for the uncertainty of outcomes

 Enable better modelling of the correlations between variables

 Test a wider range of scenarios

 Good at identifying extreme outcomes, which may not have been thought of under a deterministic scenario

 Important in assessing the impact of financial guarantees

Question 4

Steps in a deterministic model (9)

Answer 4

 Specify the purpose of the investigation

 Collect, group and modify data

 Choose the form of the model and its parameters / variables

 Ascribe values to those parameters using past experience / estimation

 Construct a model based on expected cashflows

 Test the model and correct if necessary

 Check goodness of fit against past data, modify fit if poor

 Run using estimates of future values of variables

 Sensitivity test (and maybe scenario test) using different parameter values

Question 5

Additional steps in a stochastic model

Answer 5

1. Choose a density function for each stochastic variable
2. Specify correlations between variables
3. Run model many times using a random sample from the chosen density function
4. Produce a summary of results – a distribution (eg summarised at various confidence levels