Chapter 9 - Traditional Risk Classification Flashcards

1
Q

Explain the importance of selecting proper rating variables.

A

The selection of rating variable is important to

  • Avoid adverse selection
  • Predict correctly future loss while keeping operational and legal criteria
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

explain adverse selection and the ultimate outcome if not rectified.

A

Adverse selection is when a company loses favorable risks while attracting unfavorable risks due to not properly reflecting rate differentials that competitors are capturing.

The process will continue until the compagny

  • Refine/improve is segmentation
  • Became insolvent
  • Write only high-risks and charge approriate rates
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

Describe favorable selection and alternatives to a rate change when you have identified an opportunity that could lead to favorable selection.

A

Favorable selection happen when a compagny is able to identifies a characteristic that differentiates risk that other compagnie are not using, resulting in a better segmentation.

  • Implement it in rating to improve profitability
  • “Skiming the cream”: use this information to select risks in underwriting process (writer better risks, reject worse), which will also increase profitability
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

When selecting rating variables, expain the Statistical Criteria.

A

Our rating should result in groups that are

  • Statistically significant - should estimate difference expected loss within an acceptable level of statistical confidence and should be stable over years.
  • Homogenous - groups shpuld have similar expected loss.
  • Credible - need enough data to accurately estimate expected loss.
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

When selecting rating variables, expain the Operational Criteria.

A

Our rating variable should be

  • Objectif - should have an objectif definition
  • Inexpensive to obtain - benefit of adding the variable should not outweigh the cost of obtaining it.
  • Verifiable - not subject to untruthful insureds or distribution channels.
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

When selecting rating variables, expain the Social Criteria.

A

We should consider

  • Affordability - need enough data to accurately estimate expected loss.
  • Causality - causal relationships between variable and premium.
  • Controllability - ability to control the class to which they belong by reducing their risk.
  • Privacy - must respect insured privacy.
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

When selecting rating variables, expain the Legal Criteria.

A

Above any actuarial consideration, our rating structure must follow any legal requirement.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

Explain what adjustment is necessary on Losses and Expenses when performing Traditional Risk Classification?

A
  • In practive, it is hard to allocate ULAE by class, so we often use loss & ALAE.
  • For Short-tailed - may not be necessary to trend and develop
  • For Long-tailed or when compagny is growing (shrinking) - apply trend and development factor at a aggregate level.
    • Unless there is a reason to velieve that loss or expenses vary by class.
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

Discuss the distortion related to 3 different methods of Traditional Risk Classification.

A

The Pure Premium method assume a uniform distribution accross all other rating variable, which will create material bias when it is the case.

Loss Ratio and Pure Premium Adjusted method will adjust for distributional bias/correlations only if the current relativities reflect these distributional bias/correlations.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

What is the primary weakness of the univariate classification risk method?

A

They do not accurately account for the effect other variable have on the variable being evaluated.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly