Actuarial Notes

Data Quality

By Actuarial Standards Board · 2016 · ASOP No. 23

Study guides: Exam PCPA, Exam CP-321, Exam GH-101, Exam GH-201U, Exam GH-301

The U.S. actuarial standard of practice that guides actuaries when selecting data, reviewing it, using it, or relying on data supplied by others. This revised edition of the standard on data quality (Doc. No. 185) was developed by the Data Quality Task Force of the General Committee of the Actuarial Standards Board and adopted by the ASB in December 2016; it is effective for any actuarial work product for which data were provided to or developed by the actuary on or after April 30, 2017.

On the syllabus

Transmittal Memorandum

  • The ASB first adopted ASOP No. 23 in 1993 and revised it in 2004; this revision, begun in 2014, updates its language to keep pace with practice changes such as non-traditional data sources for predictive models and legislatively mandated data submissions.
  • No significant changes were made, but wording was clarified: the defined term "comprehensive" became "sufficient", "data" now includes information derived mathematically from data, and section 3.7 (Confidentiality) was added.

1 Purpose, Scope, Cross References, and Effective Date

  • 1.1 Purpose
  • 1.2 Scope
    • Applies when selecting data, performing a review of data, using data, or relying on data supplied by others; an actuary preparing data that other actuaries will use applies it as though planning to use the data.
    • Does not apply to the generation of a wholly hypothetical data set, and does not require an audit of the data.
  • 1.3 Cross References
  • 1.4 Effective Date

2 Definitions

  • 2.1 Appropriate Data
    • Data suitable for the intended purpose of an analysis and relevant to the system or process being analyzed.
  • 2.2 Audit
    • A formal and systematic examination of data for the purpose of testing its accuracy and completeness.
  • 2.3 Data
    • Numerical, census, or classification information, or information derived mathematically from such items, but not general or qualitative information; assumptions are not data.
  • 2.4 Data Element
  • 2.5 Review
    • An examination of the obvious characteristics of data to determine if they appear reasonable and consistent for the assignment; less detailed than an audit.
  • 2.6 Sufficient
    • Containing enough data elements or records for the analysis.

3 Analysis of Issues and Recommended Practices

  • 3.1 Overview
    • Appropriate data that are accurate and complete may not be available; the actuary uses available data that allow the analysis and discloses significant known limitations.
  • 3.2 Selection of Data
    • The actuary considers the data elements desired and alternatives, and selects data considering whether they are appropriate and sufficiently current, internally consistent, reasonable given relevant external information, and sufficient; their known significant limitations; alternative data; and any sampling methods.
  • 3.3 Review of Data
    • The actuary performs a review unless it is not necessary or not practical, and discloses when none was done; a review makes a reasonable effort to determine each data element's definition and to identify questionable values or significantly inconsistent relationships, and considers comparing the data with those of a prior analysis.
  • 3.4 Use of Data
    • The actuary judges whether the data are acceptable, need enhancement, can be used with judgmental adjustments or assumptions (or with the results adjusted, such as a wider range of reasonable estimates), call for a more extensive review, or are so inadequate that the actuary obtains different data, completes the suitable parts with the principal's consent, or declines the assignment.
  • 3.5 Reliance on Data Supplied by Others
    • The accuracy and completeness of supplied data are the suppliers' responsibility; the actuary may rely on them, subject to sections 3.3 and 3.4, and discloses the reliance.
  • 3.6 Reliance on Other Information Relevant to the Use of Data
  • 3.7 Confidentiality
    • Confidential information in data is handled consistent with Precept 9 of the Code of Professional Conduct.
  • 3.8 Limitation of the Actuary's Responsibility
    • The actuary need not determine whether data are falsified or intentionally misleading, compile additional data solely to search for questionable data, or audit the data.

4 Communications and Disclosures

  • 4.1 Communication and Disclosure
    • A communication follows ASOP No. 41, Actuarial Communications (see Actuarial Communication).
    • It discloses, when relevant and material, the sources of the data, limitations from uncertainty about data quality, whether a review was performed, unresolved questionable data values, significant steps taken to improve the data, significant judgmental adjustments, results that are highly uncertain or biased, and the extent of reliance on others.

Appendix 1 Background and Current Practices

  • Informational, not part of the standard.
  • Data frequently contain errors, are not complete, and are not precisely appropriate for the intended analysis; actuaries use professional judgment on whether and how to refine data or modify the analysis, and typically do not audit data.

Appendix 2 Comments on the Exposure Draft and Responses

Related readings

  • ASOP 56 - Modeling (ASB - 2019) — refers actuaries to this standard for the data used in a model (§§3.1.5, 3.3, 4.1); both are Domain A references in the PCPA content outline

Sources