"Dynamic Claims Reserving in Non-Life Insurance: A State Space Approach with Kalman Filtering and Monte Carlo Forecasting" | ||||
| المجلة العلمية للدراسات والبحوث المالية والتجارية | ||||
| Volume 6, Issue 2 - Serial Number 1, July 2025, Page 1055-1085 PDF (1.76 MB) | ||||
| Document Type: المقالة الأصلية | ||||
| DOI: 10.21608/cfdj.2025.393750.2292 | ||||
 View on SCiNiTO | ||||
| Authors | ||||
Ahmed Samy Said El Azab  ; Raghda Ali Abdelrahman; Zahra Salah Eldin
| ||||
| Department of Insurance and Actuarial Science, Faculty of Commerce, Cairo University | ||||
| Abstract | ||||
| Accurate estimation of claims reserves is essential to maintaining the financial stability of non-life insurers, particularly in markets subject to structural volatility and evolving regulatory standards. Traditional stochastic reserving models such as the Mack model are widely used for their transparency and analytical simplicity, yet they often fall short in environments characterized by irregular reporting, dynamic settlement patterns, and macroeconomic disruption. To overcome these challenges, this research proposes a Scalar State Space Model (SSM) integrated with Kalman filtering. Applied to cumulative paid claims data from an Egyptian motor insurance portfolio, the model captures latent calendar-year effects and enables recursive reserve updating as new data becomes available. Through Monte Carlo simulation, the SSM produces full predictive distributions of future liabilities, offering a comprehensive view of reserve uncertainty. Comparative analysis against the benchmark Mack model shows that the SSM delivers more stable reserve estimates and better reflects underlying risk, especially in recent accident years where uncertainty is most pronounced. Unlike the Mack model’s reliance on fixed development patterns and independence assumptions, the SSM dynamically models time-varying processes and structural shocks. The results highlight the SSM’s advantages in adaptability, robustness, and regulatory alignment, making it a compelling alternative for insurers operating under IFRS 17 and similar solvency-focused regimes. These findings advocate for the integration of dynamic stochastic reserving methods into the actuarial toolkit, particularly in emerging markets facing data volatility and structural transformation. | ||||
| Keywords | ||||
| State Space Model; Kalman Filter; Mack Model; Monte Carlo Simulation; Stochastic Reserving | ||||
|
Statistics Article View: 53 PDF Download: 51 |
||||
