Attention: Due to COVID-19 restrictions, seat capacity at Prometric testing centers continues to be affected. If you are unable to find a seat for your exam, click here to see your options.

SOA - Course 7 Learning Objectives

Course 7 Learning Objects

Course 7: Applied Modeling
Learning Objectives

Overall Objectives:

The candidate must demonstrate the ability to appropriately apply the modeling process in order to support recommendations and/or facilitate business decision-making.  Further, the candidate must effectively communicate the findings and/or implications of his/her model to technical and non–technical audiences.  The emphasis of the course is not on specific modeling techniques but on modeling process, business problem solving, and communication.  At the seminar, technical knowledge of a limited number of models and/or modeling techniques will provide the context for assessing the primary objectives.

Within the context of these overall objectives, the candidate must demonstrate knowledge and capability in the following areas:

  1. The Context of Modeling:
    The candidate shall be able to:

    1. Define a model
    2. Define an actuarial model
    3. Demonstrate a general understanding of the modeling techniques used in actuarial practice such as, but not limited to, survival models, credibility models, risk theory models, ruin theory models, option pricing models, cash flow and cash flow testing models, and nontraditional models by
      1. Defining the general characteristics of each modeling technique
      2. Describing the characteristics of the data, assumptions and/or input required to specify a unique model
      3. Describing the characteristics of the output of each modeling technique
      4. Recognizing alternative modeling techniques that may be appropriate for solving a particular business problem

    4. Explain the modeling process, including the feedback loop
    5. Recognize when a modeling approach is appropriate or inappropriate.  When a modeling approach is appropriate, recognize when a simplistic approach may be sufficient

    6. Apply principles underlying models, by
      1. Defining principles common to all models
      2. Creating models that apply the principles appropriately
      3. Recognizing when principles have been violated and if any such violations have material effect on the solution to a business problem
      4. Adjusting a model or the output of a model to correct for material violations of principles

    7. Identify and describe limitations of specific applications of the modeling process.

    8. Identify and describe sources of error in the modeling process, including:
      1. Process error (pure risk)
      2. Statistical estimation error
      3. Model selection error
      4. Model versus the universe, and
      5. Assumption error, including explicit and implicit assumptions about the future environment

  2. Model Design, Selection and Set-Up:
    The candidate shall be able to:

    1. Select and apply model(s) appropriate to solving business problems
    2. Justify his/her model selection(s)
    3. Calculate and explain potential errors in the model(s) selected
    4. Select and justify reasonable and appropriate assumptions to the selected underlying model(s)
    5. Select and justify the parameters of any parametric model(s) selected
    6. Explain the explicit and implicit advantages and limitations of alternative models
    7. Explain how the model(s) selected was influenced by data quality and accessibility, available resources and output requirements
    8. Explain how professional and regulatory requirements affect the model(s) selected
    9. Explain explicit and implicit assumptions of the model(s) selected
    10. Assess model usefulness using a variety of techniques including sensitivity analysis

  3.  Input Data Selection and Analysis:
    The candidate shall be able to:

    1. Assess the quality and relevance of a given data set for solving business problems
    2. Evaluate and assess the effect of data quality on the solution to a specific business problem
    3. Balance data quality, accessibility, credibility and relevance when selecting the data needed to solve a business problem
    4. Identify, if possible, alternate data sources for solving a business problem
      1. Explain the variety, reliability and availability of data from each source

  4.  Analysis of Results:
    The candidate shall be able to:

    1. Assess the reasonableness of the results of the modeling process
    2. Measure the sensitivity of output to changes in the input, model and model parameters.  Assess the effect of sensitivity on the usefulness of the results.
    3. Integrate the results of several models together
    4. Draw conclusions and/or make recommendations that support business decision-making
    5. Recognize the useful life of a model, its input and its assumptions

  5. Communicating the modeling process:
    The candidate shall be able to:

    1. Effectively communicate his/her conclusions, model, and limitations to technical and non–technical audiences alike.  The communications shall recognize:
      1. The nature of the audience
      2. Professional requirements (Standards of Practice)
      3. Regulatory requirements

    2. Select appropriate format and medium for his/her communications
    3. Create and maintain sufficient documentation to meet professional standards