Insurance companies need accurate life tables representing the mortality of their own portfolios, products or segments. Those tables can be used for pricing, valuation and profitability purposes. Calibrating a portfolio-specific lifetable directly on the portfolio historical data is however often impossible due to the lack of data.
The market practice is to fit a lifetable on a “general population” portfolio (typically the data of a country) and then to adapt it to fit the “specific population” mortality (typically an insurance portfolio). Such adaptation is performed thanks to an adverse selection model. The Reacfin tool relies on various methods for carrying out such adaptations, such as the Generalized Linear Models (GLM) or the Local Fitting Model (LFM). It also enables the user to set a broad range of parameters for the aforementioned methods. In order to facilitate further analysis the tool outputs are available for download.