New Research RFP Seeks Insight from Researchers Outside of the Insurance Industry on Forecasting Future Loss Payments from Policies Sold in the Past
The Casualty Actuarial Society (CAS) invites researchers outside the insurance industry, with expertise in modeling longitudinal data, to participate in a property-casualty insurance case study. The case study involves forecasting loss payments to cover claims from insurance policies sold in the past using data sets to be provided by the CAS.
Up to $50,000 in funding is available for a study that will be presented as a peer-reviewed research paper. The study will be based on examples typical of the loss payment forecasting assignments CAS members face in practice.
Research Problem Description
A common analysis task for members of the CAS is to forecast future loss payments grouped by exposure periods and coverage. Actuaries use the term loss reserving for that forecasting exercise. Loss reserves are the estimate of the total amount of future payments that will be made to claimants to satisfy the terms of the policies sold by the company. The forecasts are used to establish liabilities the company faces to pay claims for policies sold in the past.
The data sets for the case study will be in the form of a longitudinal data set(s). A longitudinal data set captures observations on the value to be forecast or evaluated from the subjects of the study at regular intervals. Longitudinal data sets are commonly found in pharmacology studies where the effect of a drug is evaluated on patients over time or in agronomy where the effect of different fertilizers on crop growth is evaluated over time. In the CAS-provided dataset, the subjects are groups of policies split by exposure periods and the coverage offered by those policies. The observations gathered at regular time intervals are statistics from claims covered under the exposure periods for which the policies were in effect.
The data sets furnished for the case study will be generated via simulation and will be designed to replicate selected modeling problems that are commonly faced by members of the CAS.
The history in the longitudinal data sets for claim activity can be affected by a variety of changes in:
- General inflation.
- Repair or replacement cost unique to the coverage granted under the policies.
- Laws governing claim payment practices.
- Internal claims handling practices.
- Insured population due to marketing or pricing changes.
Claim payments related to incidents covered during the policy effective period can extend for several years into the future. Future claim payments may be subject to different inflation effects than those shown in the longitudinal data which introduces an additional layer of uncertainty when forecasting.
Research Requirements
Case study deliverables include:
- Explanation of the modeling technique selection.
- Test statistics for evaluation of the results.
- Examples of how to present the results to a non-technical audience.
- The open-source code in that was used to generate the results.
- Forecasts that give the distribution of future loss payments.
If researchers find that some form of tempering or regularization is necessary when modeling (actuaries call that credibility weighting), an explanation is required. If the researchers find that there is correlation in the claim statistics between time periods that should be incorporated in the forecasting work, then an explanation of how that was measured and then incorporated in the forecast is required.
We are open to code that uses either R or Python, but we expect the packages used in the study will be vetted by an organization like The Comprehensive R Archival Network for R packages, CRAN, or PyPI (an index for Python packages) and that the code used to build the modeling results will have comments explaining the modeling sequence. The paper also requires a bibliography so actuaries can obtain a broader background in the techniques used in the paper’s case study.
We are looking for submissions from researchers outside of the P&C industry to give a fresh perspective on how to estimate future loss payments. A clear explanation of why a given technique is viewed as the best choice compared to other techniques is as important as producing plausible modeling results. If the path taken in analyzing the data sets furnished by the CAS is like other studies that the researchers have performed in areas like agronomy of pharmacology, pointing out what types of studies are similar in nature to the problem of estimating future loss payments would be essential.
Submitting Proposals
Proposals should include:
- A clear outline of the workplan and timeline.
- Researchers resume detailing qualifications.
- Cost estimates (not exceeding $50,000) (Note: A portion of the grant can be used to cover usual and customary travel expenses to present the paper at a CAS-sponsored seminar or meeting.
The CAS contract will be awarded to the respondent who, in the judgment of the CAS Reserves Working Group and based on the written proposal, is deemed to be best able to perform the work as specified. If the group determines that no proposal meets the requirements of the RFP, no contract will be awarded. Interested parties are welcome to submit questions about the RFP before the proposal deadline of Monday, October 27, 2025.
Proposals and questions should be sent to Elizabeth Smith, CAS Director of Publications and Research, by October 27, 2025. Proposal submissions should write “Longitudinal Reserving RFP Proposal” in the subject line.
Receipt of proposals will be acknowledged promptly. Respondents who are not awarded the contract will be informed by Monday, December 1, 2025.
Review of Paper
The research paper will be assigned to three or more members of the Reserves Working Group to serve as reviewers.
These reviewers will read drafts, provide feedback and help the researcher(s) develop a final draft. The goal of this review process is to help researchers produce a final paper suitable for publication of the case study in a CAS publication.
The researchers may elect to produce an additional paper related to the case studies that distills the material into generalized, mathematically oriented guidelines for forecasting loss payments that would be eligible for publication in the CAS’s refereed journal, Variance, pending review by that journal’s editorial staff.
Compensation
Compensation to researchers will be commensurate with the scope of the proposed project subject to an overall cap of $50,000.
Authors will be required to upload their final additional paper electronically in the CAS’s ScholarOne system and must report on the use of artificial intelligence, if any, used to produce research.
Travel costs to present the results of the project at a CAS seminar will be covered subject to the constraints of the overall total cost cap of $50,000.
Presentation, Ownership and Publication of Report
As a condition of selection, the CAS requires the right to publish the research paper in any of our publications. The research authors may publish the results of their work in other publications but will acknowledge in any publication outside of those by the CAS the role that the CAS played in supporting the research effort. The selected researcher/research team must sign a formal research agreement assigning all such rights to the CAS. In any publication of the report, the researcher(s) will receive appropriate authorship credit. The CAS may publish the report in its entirety, or any sections thereof, in any format and medium, including, but not limited to CAS publications, and electronic versions on its website or physical storage media. To aid research adoption, the final work product’s code and data will also be placed in the CAS’s GitHub repository, https://github.com/casact, under the MPL2.0 license.
The researcher(s) should make every effort to be available to present the report at a CAS meeting or seminar.
Timeline
Monday, October 27, 2025 |
Proposals Due |
|
Monday, December 1, 2025 |
Authors Notified |
|
Monday, June 15, 2026 |
Completed Project Deadline |
About the Casualty Actuarial Society (CAS)
The CAS was organized in 1914 as a professional society for the promotion of actuarial and statistical science as applied to insurance other than life insurance, such as automobile, liability other than automobile, workers compensation, fire, homeowners, commercial multiple peril, and others. Such promotion is accomplished by communication with those affected by insurance, presentation and discussion of papers, attendance at seminars and workshops, collection of a library, research, and other means. The membership of the CAS includes over 10,000 actuaries worldwide, employed by insurance companies, industry advisory organizations, national brokers, accounting firms, educational institutions, state insurance departments, the federal government and independent consultants.
