Practical Tools for Loss Development Using Non-Aggregated Methods

Details

Supervisors

Denuit, Michel ; Sime, Achille

Faculty

Faculté des sciences

Degree label

Master [120] en sciences actuarielles, à finalité spécialisée

Abstract

Insurance is a reverse economic cycle in which the insurer first sets its selling price (premium) before knowing its cost (claim amount). This creates a real need for (re)insurers to establish technical provisions to honor these commitments. Loss reserving is typically done using "aggregated" methods. These methods involve adding up the data by accident year or policy period and evaluation period. These methods have several advantages, including but not limited to, ease of implementation or communication of the results. However, aggregating claims may exclude valuable information that could improve the accuracy of estimations, particularly for late-reported claims such as reporting delays. This thesis focuses on evaluating insurance loss reserves, particularly the Incurred But Not Reported (IBNR) reserves using Non-Aggregated methods, proposing enhancements to the valuation tool designed by SL FINANCIAL, Inc. (“SL FINANCIAL” or the “Company”) using Excel VBA code, and comparing the results to those obtained using traditional methods. The traditional methods considered include the Chain Ladder, Bornhuetter-Ferguson, Cape Cope, and Berquist-Sherman methods. The findings of this study provide valuable insights for insurance companies seeking to improve their IBNR reserve evaluation processes. These methods allow for separate estimates of Incurred But Not Yet Reported (IBNYR) or pure IBNR and Incurred But Not Enough Reported (IBNER), leading to superior decision-making.