One-Shot Individual Claims Reserving

Paper: arXiv 2603.11660

Authors: Ronald Richman, Mario V. Wüthrich

Abstract

Individual claims reserving has not yet become established in actuarial practice. We attribute this to the absence of a satisfactory methodology: existing approaches tend to be either overly complex or insufficiently flexible and robust for practical use. Building on the classical chain-ladder (CL) method, we introduced a new perspective on individual claims reserving in Richman and Wüthrich [arXiv:2602.15385]. This manuscript has sparked considerable discussion within the actuarial community. The aim of the present paper is to continue and deepen that discussion, with the ultimate goal of advancing toward a new standard for micro-level reserving.

Complexity vs Empirical Score

• Math Complexity: 5.5/10
• Empirical Rigor: 7.5/10
• Quadrant: Holy Grail — high math complexity, high empirical rigor

Why this score: This paper bridges the gap between classical actuarial chain-ladder methods and modern machine learning frameworks for micro-level claims reserving. It demonstrates strong practical relevance and empirical execution using linear models, neural networks, and transformers, while responding thoroughly to community feedback on arXiv.

Research Flowchart

  flowchart TD
    A["Research Goal: Advance a new standard for micro-level reserving"] --> B["Data Inputs: Individual Claims Data & Classical Chain-Ladder Principles"]
    B --> C["Methodology: Develop One-Shot Individual Claims Reserving Framework"]
    C --> D["Computational Process: Extend Richman & Wüthrich Model for Micro-Level Reserving"]
    D --> E["Key Findings: Provides a Flexible, Robust, and Practical Alternative to Complex Models"]