PRICING CATASTROPHIC BOND PREMIUMS USING EXTREME VALUE THEORY: CASE OF BUSHFIRE IN HEALESVILLE, AUSTRALIA

• Main Author: Christella Hidayat, Jessica
• Format: Final Project
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/42417
• Institution: Institut Teknologi Bandung

One way to cover financial losses caused by natural disasters is through insurance. As an alternative way to finance the risks of financial losses, a reinsurance company which covers losses due to catastrophic events may use a Special Purpose Vehicle (SPV) to sell a Catastrophe Bond (CAT-Bond) to potential investors. This final project discussed the topic of pricing the premium of a catastrophe bond using the Extreme Value Theory with the Peaks Over Threshold (POT) approach. The POT approach may be argued as being more effective in modeling extreme values compare to the Block Maxima approach. Through the POT approach, the excess data over a certain threshold u may be assumed to follow a Generalized Pareto Distribution (GPD) with the shape parameter and the scale parameter . In this final project, a case study is given in which the underlying catastrophic event is a bushfire disaster in Healesville, Australia; and the trigger is the level of precipitation. The determination of the premium of the corresponding CAT-Bond is carried out using a financial loss premium principle involving a financial market; in this case, the S&P500. The Extreme Value Theory is applied to the precipitation data in Healesville, Australia. It is obtained that, for a threshold of u = 0:975, the estimates of the parameters are ^ = ????0:136463 and ^ = 0:314456 with l = 3:279334. Furthermore, the analysis of different values of triggers and the corresponding premiums of the CAT-Bonds are carried out for two cases: a risk-averse investor and a risk-lover investor.