La Riservazione Sinistri nell'Assicurazione Danni: Un'Analisi Comparativa dalle Tecniche Classiche a quelle Avanzate

In the financial management of insurance companies, the process of claims reserving is of critical importance. It determines the necessary amount to set aside as reserves in order to guarantee the solvency of the organisation. For this purpose, in practice actuaries have to select the appropriate reserving methodologies checking the underlying assumptions. This thesis aims to provide an overview of the topic, starting from the regulatory framework in force in Italy and in Europe. This allows to understand the importance of compliance and the implications for financial stability. After a brief introduction of fundamental basic concepts, the thesis delves into the various methodologies employed in claims reserving, highlighting both traditional and advanced approaches, focusing on the most widely used ones. Traditional methods include the Chain - Ladder, Bornhuetter - Ferguson and the Fisher - Lange methods, each one presented with its own assumptions, properties and applicability depending on the data and context. On the other hand, advanced methods such as the Mack model and Bootstrap techniques are discussed. The Mack model, a stochastic approach, allows for the estimation of the variability around the reserve estimates, providing a measure of the uncertainty. The Bootstrap method, widely used for its flexibility, involves resampling techniques to create multiple simulated datasets, which help in understanding the distribution of possible outcomes. Finally, the thesis includes a data analysis section in which the methods covered are applied to real-world data from six companies operating in different lines of business. The results are presented, the performance of the models is compared and possible further research is suggested.