Modelling

Question 1

Steps for modeling

Answer 1

Specify the purpose and key features of the model
Obtain and adjust the data
Set the parameters/assumptions, including dynamic links
Construct the model cash flows
Check the accuracy and fit of the model, and amend if necessary
Run the model as many times as required
Output and summarize the result

Question 2

What is sensitivity analysis and why is it useful

Answer 2

Sensitivity analysis is the re-running of a model with different, but feasible parameter values that produce alternative results, helping to illustrate potential deviations and the financial significance of key parameters
Understanding this exposure can help the manager with allocation

Question 3

Define deterministic modeling

Answer 3

The parameters are fixed at outset
Results of running model is a single outcome
Potential variability is assessed by sensitivity analysis and scenario testing

Question 4

Define stochastic modeling

Answer 4

At least one parameter is estimated
By assigning it a probability distribution with value of stochastic parameters randomly selected from their distribution on each run
Outcome is range of values

Question 5

Merits of a deterministic model

Answer 5

More readily explainable
Concept of variables not easy
easier to understand scenarios
Cheaper and easier to design
Quicker to run

Question 6

Merits of stochastic model

Answer 6

Wider range of scenarios
Higher quality result
Complex programming and long run times
Depends on parameters used
Important to assess impact of guarantees or investment mismatching

Question 7

idea operational requirements for any model

Answer 7

the model should be adequately documented

workings of the model should be easy to appreciate and coomunicate

should exhibit sensible joint bahaviour of variables

output of the model should be capable for independent verification for reasonableness

the model must not be overly complex or time-consuming and expensive to run

the model should be capable of development and refinement

there should be methods in place to facilitate the testing of the model

Question 8

Issues to condier when deciding whether to use determinstic or stochastic model

Answer 8

deterministic model:
could test stresses using variations in assumptions
easier to test effects of defined scenarios

stochastic model:
more complex to build
difficult to explain to a wider audience
more model risk
harder to parameterise - greater parameter risk
objective - incorporate allowance for volatilities in asset values and uncertainty in claims experience
allow correlations between assets and liabilities
allows construction of probability distribution - prob of ruin and capital required to avoid ruin