Application Of Value At Risk (VaR) Models On Insurance Claim Data

Abstract

Measuring market risk is one of the main problems for the finance and insurance companies. As a measure of possible losses of an investment, Value at Risk (VaR) is widely used risk metric in financial area. After the financial crisis, the regularities force the financial institutions to report their VaR measure in their financial reports. In this paper, we represent the application of various VaR models on insurance claim data by using related R packages. We mention the importance of reporting VaR for insurance companies to be able to compare their effectiveness and compete in their area.

Insurance data, which includes a large number of small losses and a lower number of very large losses, is generally non-normally distributed. Therefore we also try to attract the attention to importance of fitting the distributions in terms of accuracy of measurement.