C18 Data(2): Quality Flashcards
When data is required for a number of tasks, what is the key principle in its provisioning
Overriding principle concerning uses of data:
- One single, integrated data system so that the data used for different applications is consistent
(Note that this is not always achieved in practice but is easier to ensure in a smaller organisation)
List the uses of data
Uses of data : AIR SPAMMER
Accounts
Investment monitoring
Risk management
Statutory returns
Pricing
Administration
Marketing
Management information
Experience analysis and statistics
Reserving/Provisioning
Causes of poor data quality and quantity
- Poor management control of data recording and checking
- Poor design of data systems
Ensuring good quality data from the proposal and claim forms
- Well designed and unambiguous questions
- Forms designed so that information can be easily analysed, and cross checks made between two sources
- Questions in same order as input into admin systems so applications and claims can be processed quickly
- Rating factors readily identifiable
- Results of underwriting process added to proposal form
List use of proposal form when assessing claims
Need for proposal forms at time of claim:
- To help check the validity of the claim
- To update policy information, e.g. policyholder has died
State the Importance or retaining past policy and claims records
Importance or retaining past policy and claims records as well as current risks
- When analyzing past experience to help set future assumptions, several past years’ worth of data are often needed to give sufficient volume of data, or to identify trends
Outline data checks that might be done on the data
Reconcile:
1. Total number of members/policies and changes in membership/policies using previous data and movement data
2. Total benefit amounts and premiums and changes in them, using previous data and movement data
3. Shareholdings and the start and end of the period, adjusted for sales and purchases and bonus issues
Cross-Check:
4. Movement data against accounting data e.g. benefit payments
5. Membership data against accounting data
6. Asset data against accounting data
7. Asset ownership records against custodian’s records
Reasonableness checks
8. Average sum assured or premium looks sensible for class of business and against previous data
9. Unusual values
10. Missing data
Spot checks
11. Individual member or policy records
12. Individual assets and liabilities exist on a given date
13. Correct value of an asset or liability has been recorded
14. Carry out a full deed audit for certain assets
Data Issues for Employee Benefit Schemes
Problems with data for employee benefit schemes:
1. Actuary may not have full control over data, as it is provided by the sponsor.
2. It may be poor quality or summarized data
3. Particularly important to check this type of data
Data requirements for Employee Benefit Schemes
Data requirements for Employee Benefit Schemes
1. Place a value on the benefit entitlements on an individual
2. Data should have all info that is financially significant to the level and timing of future benefits (e.g. age)
Data sources for Employee Benefit Schemes
Sources of data:
1. Membership data (sufficiently detailed) on individuals who are currently receiving benefits and those who are entitled to in the future
2. Data from previous valuation for reconciliation with current data
3. Accounting data for checks on assets, benefit outgo and contribution income
4. Full listing of the actual assets held to check whether they are permitted
List 5 assertions related to the data that an actuary should check
An actuary will have to make and check certain assertions about data
1. A liability or asset exists on a given data
2. A liability is held or an asset is owned on a given data
3. Time of the event and the associated income or expenditure are allocated to the correct accounting period
4. Data is complete i.e. no unrecorded liabilities, assets or events
5. Appropriate value of an asset or liability has been recorded
Describe the drawbacks of using summarised data for valuation purposes
Problems:
1. Reliability of valuation will be reduced as full validation of the data is impossible
2. May miss significant differences in benefits that have been groups together
3. Changes in mix of members may remain unidentified
4. Not useful for valuing individual options and guarantees
5. Only suitable if such inaccuracies are recognised by users of the results
Give two examples of lack of ideal data.
- Data has not been captured at a sufficient detail e.g. member cat in benefit scheme, limited data on insurer’s database
- Insufficient data to provide credible result e.g. new product or Target market, benefit scheme size small for mort exp
Problems with using industry data
Problems with using industry data DR DONEQ
1. Detail insufficient
2. Risk factors coded in different ways
3. Difference in target market, underwriting, terms, geographical areas, sales channels, rating factors
4. Out of date
5. Not everyone contributes
6. Errors
7. Quality depends on that of contributors
Reasons why industry data is not directly comparable
Reasons why industry data is not directly comparable:
- Different geographical or socio-economic markets
- Different policies (i.e. cover, terms and conditions)
- Different sales methods
- Different underwriting and claims settlement processes
- Different nature of data stored
- Different coding of risk factors, e.g. definition of a smoker
