Ch 19: Data

Question 1

Define personal data

Answer 1

Personal data is information that relates to an individual which would allow that individual to be identified, or where the data combined with other information could allow the individual to be identified.

Question 2

8 Principles which must be followed when processing personal data.

Answer 2

Personal data must:
1) Be processed fairly and lawfully.
2) Be obtained and processed for specific purposes.
3) Be adequate, relevant and not excessive for the purposes concerned.
4) Be accurate, and where necessary, kept up to date.
5) Not be kept longer than necessary for the purposes concerned.
6) Be processed in accordance with the individual’s rights.
7) Be processed securely.
8) Not be processed to another company or country unless that party ensures an adequate level of protection.

Question 3

Examples of what might count as ‘sensitive personal data’

Answer 3

Sensitive personal data can include information related to:
1) Racial or ethnic origin
2) Political opinions
3) Religious or other beliefs
4) Membership of trade unions
5) Physical or mental health or condition
6) Sexual life
7) Convictions, proceedings and criminal acts.

Question 4

Give examples of circumstances when sensitive personal information may be legitimately processed

Answer 4

1) The data subject has given explicit consent.
2) It is required by law for employment purposes.
3) It is needed in order to protect the vital interests of the individual or another person.
4) It is needed in connection with the administration of justice or legal proceedings.

Question 5

State 3 characteristics of ‘big data’

Answer 5

Big data can be characterised by:
1) The data sets being very large
2) Data is brought together from different sources
3) Data can be analyzed very quickly, for example in real time

Question 6

State 4 risks to a company not having adequate data governance procedures

Answer 6

1) Legal and regulatory non-compliance
2) Inability to rely on data for decision making
3) Reputational issues, leading to loss of business
4) Incurring additional costs such as fines and legal costs

Question 7

Define ‘data governance’ and list the guidelines that a data governance policy may cover.

Answer 7

Data governance - the overall management of the availability, usability, security and integrity of data employed in an organisation

A data governance policy will set out guidelines with regard to:
1) The specific roles and responsibilities of individuals in the organisation with regards to data.
2) How an organisation will capture, analyze and process data.
3) Issues with respect to data security and privacy.
4) The controls that will be put in place to ensure that the required data standards are applied.
5) How the adequacy of controls will be monitored on an ongoing basis with respect to the data usability, accessibility, integrity and security.
6) Ensuring that the relevant legal and regulatory requirements in relation to data management are met by the organisation.

Question 8

List the main sources of data.

Answer 8

TRAINERS

• Tables
• Reinsurers
• Abroad (data from overseas contracts)
• Industry data
• National Statistics
• Experience investigations on the existing contract
• Regulatory reports and company accounts
• Similar contracts

Question 9

What is the overriding principle in relation to all the different uses of data?

Answer 9

There should be one single, integrated data system so that the data used for different applications is consistent.

Question 10

Define algorithmic trading

Answer 10

This is a form of trading that involves buying and selling financial securities electronically to capitalize on price discrepancies for the same stock or asset in different markets

Question 11

Explain the risk of algorithmic trading (4)

Answer 11

1) Errors in the algorithm or data used to parameterize the model, leading to losses.
2) The algorithm may not operate properly in adverse conditions
3) In very turbulent conditions, trading in individual stocks or markets may be suspended before algorithmic trade can be completed
4) Possible impacts on the financial system

Question 12

List the key risks associated with using data (7)

Answer 12

1) Data are inaccurate or incomplete, leading to erroneous results or conclusions
2) Data are not credible
3) Data are not sufficiently relevant to the intended purpose
4) Past data do not reflect what will happen in the future
5) Chosen data groups are not optimal
6) Data not available in an appropriate form for the intended purpose
7) Lack of confidence in the data leads to a lack of confidence in the results obtained from using it

Question 13

What 2 main factors cause data to be of poor quality and quantity?

Answer 13

1) Poor management and control of data recording and checking
2) Poor design of data systems

Question 14

How can good quality data be ensured from an insurance proposal and claims form? (7)

Answer 14

1) Questions should be well designed and unambiguous so that full information is given and so that applications/claims can be easily processed
2) Use questions with quantitative or tick-box answers wherever possible
3) Questions should be in the same order as the input into the administration systems, for quick processing of applications/claims
4) Ask the policyholder to verify key information
5) All rating factors should be readily identifiable so that the composition of the final premium can be determined
6) Underwriting results should be added to the proposal form
7) Forms should be designed so that information can be easily analyzed, and cross-checks made between the two sources.

Question 15

Why is it important, at the time of the claim, to have access to the information given on the proposal form? (2)

Answer 15

1) To check the validity of the claim
2) To update policy information

Question 16

Why is it important that the insurance company retains a past history of policy and claims records?

Answer 16

When an insurance company analyses past experience in order to help set future assumptions, several years’ worth of data are often needed in order to give a sufficient volume of data, or to identify risks.

Question 17

What is the key problem with data for employee benefit schemes?

Answer 17

The actuary does not have full control over the data, as it is provided by the sponsor.

The consequences of this may be poor quality or summarised data

NOTE: It is therefore particularly important to validate this type of data.

Question 18

What four sources of data are useful in order to conduct a valuation of a benefit scheme?

Answer 18

1) Membership data on individuals who are currently receiving benefits and those who are entitled to in the future.
2) Data from the previous valuation for reconciliation with current data to help validate the current data.
3) Accounting data for information on asset values, benefit outgo and contribution income to help check other sources of data or in setting assumptions.
4) A full listing of the actual assets held to enable an accurate valuation of assets and to check whether they are permitted by regulation or subject to regulatory restrictions.

Question 19

Give examples of reconciliation checks that can be performed on data (4)

Answer 19

1) Reconciling the total number of members/policies and changes in membership/policies using previous data and movement data.
2) Reconciling the total benefit amounts and premiums and changes in them, using previous data and movement data.
3) Where assets are held by a third party, reconciliation between the beneficial owner’s and custodian’s records.
4) Reconciling shareholding at the start and end on the period, adjusted for sales and purchases, and bonus issues.

Question 20

Give examples of cross-checks that can be performed on data (4)

Answer 20

1) Checking movement data against accounting data, e.g., benefit payments
2) Checking membership data against accounting data, e.g., contributions
3) Checking asset data against accounting data, e.g., investment returns
4) Full deed audit, for example, checking title deeds to large real property assets.g

Question 21

Give examples of reasonableness checks that can be performed on data (3)

Answer 21

1) Checking whether the average sum assured or premium looks sensible for class of business
2) Checking whether the average sum assured or premium against previous data.
3) Checking for unusual values, impossible dates or missing records.

Question 22

Reasons why industry data is not directly comparable (6)

Answer 22

1) Different geographical and socio-economic markets
2) Different politics
3) Different sales methods
4) Different practices, e.g., underwriting and claims settlement processes
5) Different nature of data stored
6) Different coding of risk factors

Question 22

Give examples of spot checks that can be performed on data (3)

Answer 22

1) Random checking of individual member or policy data
2) Checking if individual assets or liabilities exist/are held on a given date.
3) Checking that the correct value of an asset or liability has been recorded.

Question 22

4 Other problems with industry data:

Answer 22

1) Less detailed and flexible than internal data
2) More out-of-date than internal data
3) Data quality depends on the quality of the data systems of all its contributors
4) Not all organizations contribute, and those who do may not be representative of the market.

Question 23

Outline 3 problems when using summarised data

Answer 23

1) The reliability of the valuation will be reduced, as full valuation of the data will be impossible
2) Summarised data may miss significant differences between the nature of the benefits that have been grouped together
3) Summarised data cannot be used to value options and guarantees that apply at an individual level

Question 24

What is risk classification and what is its main aim?

Answer 24

Risk classification - a tool for analyzing a portfolio of prospective risks by their risk characteristics, such that each subgroup of risks represents a homogeneous body of risk.

The main aim of risk classification is to split data into homogeneous groups so that the experience of each group is more stable, and data can be more accurately used, for example, to set premiums

Question 25

When is data ‘consistent’?

Answer 25

Consistent means that when comparing the experience of one group of policyholders with another, say, the data used as a basis for the calculations of each group should be:
- Similar
- Preferably extracted from the same source
- Grouped according to the same criteria
- Equal in terms of reliability