ERM Chapter 21

Question 1

What types of risks are associated with modelling?

Answer 1

• Model risk is that arising from the use of inappropriate or inaccurate models when assessing or managing risks
• Parameter risk refers to the use of inappropriate or inaccurate parameters or assumptions within such models

Question 2

Why might stochastic models be used above deterministic models?

Answer 2

• As the number of observations increases, so too does the certainty in the model, its parameters and its predictions. Stochastic models acknowledge the presence of this uncertainty by producing a range of results, rather than a single deterministic answer.

Question 3

Describe how model parameters might be simulated?

Answer 3

• instead of using a static set of parameter to carry out a simulation, dynamic parameters can be used instead.:
  > fit a model using T data points
  > simulate T data points using the model
  > re-fit the model to the simulated data points
  > record the parameter values
  > repeat process a large number of times, starting with the original data set every time

Question 4

What are the three assumptions that can be made in relation to choice of model?

Answer 4

1. The true model or class of models is known.
2. The model used is a simplification of a known, more complex reality.
3. The model used is an approximation to an unknown, more complex reality.

Question 5

Why may wrong models be used when approximating an unknown, more complex reality?

Answer 5

• inappropriate projection of past trends into the future, perhaps due to:
  > errors in historical data
  > incomplete data
  > heterogeneity in the data, where underlying drivers and their dependencies are not known or not projected separately
• the selection of an inappropriate underlying distribution, perhaps due to:
  > insufficient data
  > not investigating a range of alternative candidate distributions
• the number of parameter chosen without reference to:
  > the need to avoid over-simplification and the risk of implicit assumptions
  > the principal of parsimony, which states where there is a choice of fitted models, the optimal selection is the one with the fewest parameters.

Question 6

What is the prime objective in building a model?

Answer 6

• To enable the actuary or risk manager to give an organisation appropriate advice so that it can manage its risks in a sound financial way.
• Models can also be used to both assist in the day-to-day running of the company and also to provide checks and controls on the business.

Question 7

Describe the use of model points.

Answer 7

• Grouping of a very large number of individual data points into a manageable number of relatively homogeneous groups.
• Groupings are made in such a way that each policy in a group is expected to produce similar results when the model is run.
• Individual policies can they be run through the group, and the result scaled up to represent the total set of policies in the group

Question 8

What are the main requirements for a generic actuarial model.

Answer 8

• model must be valid and sufficiently rigorous for the purposes to which it will be put, and adequately documented
• model points chosen to reflect adequate distribution of the business being modelled
• components of the model must allow for all features of the business being modelled that could significantly affect the advice being given
• input parameter values should be appropriate to the business being modelled and take into account special features of the company and economic and business environment in which its operating
• workings of the model should be easy to appreciate and communicate. results should be displayed clearly
• outputs should be capable of independent verification for reasonableness and should be readily communicable to those to whom advice will be given
• model should be capable of subsequent development and refinement - nothing complex can be successfully designed and built in a single attempt
• model must not be overly complex so either the results become to difficult to interpret or communicate or the model becomes too long or expensive to run, unless required. important to avoid the impression that everything can be modelled.
• model should be amenable to an analysis of the impact of parameter uncertainty or incorrectly specified parameter values
• model should exhibit behaviour in simulations that is consistent with the past.
• any shortcomings of the model should be clearly stated

Question 9

Why might an organisation build a model for ERM decision-making?

Answer 9

• pricing products or services
• assessment of economic value of the company
• estimation of the possible volatility of future profits and earnings
• determination of capital adequacy requirement, including regulatory requirements and internal economic capital assessments
• projection of future capital or solvency position
• assessment of the effect of RM and mitigation techniques on both profits and capital requirements
• assessment of the effect of other strategic decisions
• evaluation of projects

Question 10

What are the steps in developing and applying a model?

Answer 10

• specify the purpose of the investigation
• collect data and group and/or modify as necessary
• choose the form of the model, identifying parameters and variables
• estimate the required parameters and correlations between them
• check the goodness of fit and attempt a different model if unacceptable
• ensure the model is able to project all required cashflows and other outputs, including interactions between them
• run the model using the selected estimated variables
• output the results in an appropriate format
• assess the sensitivity of results to different deterministic variable values
• for cashflows we must also determine the time period over which to project - more frequent cashflows will improve reliability but increase time and cost

Question 11

What corporate decision-making can be made using models?

Answer 11

• compare outputs to risk appetite and risk policy of the organisation
• decisions can be made based on economic value added (EVA), which is the present value of all future shareholder profits
• models might identify a number of alternative strategies on the efficient frontier. Decision-makers will have to decide on their preferred option from these alternatives
• decisions should be made based on qualitative and quantitative aspects. Relying on only quantitative models can be dangerous, particularly where the limitations of the data, parameters or model used are not fully understood or appreciated