| description abstract | Natural disasters have resulted in record losses for the last 50 years. Decision makers are rightly concerned about the vulnerability of their economies for which they have to make, under uncertainty, complex investment and policy choices. The challenge with estimating losses of extreme weather events (e.g., windstorms) is that it may take thousands of years to capture statistically meaningful events. Thus, stochastic simulation is usually incorporated in insurance pricing models. This paper proposes a risk management model to price insurance premiums for windstorm-susceptible developments. The model is applied within the regional context of Mississippi where the total windstorm events recorded between June 30, 1950 and June 30, 2010 was 1,988 and resulted in $1.875 billion losses. Mississippi was divided into three regions pursuant to the wind speed contours developed by the International Building Code. The historic data set was bootstrapped through imposing correlation to generate a data set of 5,000 events in each division. The simulated data set was modeled by using a Monte Carlo simulation on the basis of the options theory to price the fair-valued premium for windstorm insurance. A $250,000 home can be insured at 100% coverage for an annual premium of $8,200, $4,350, and $2,975 at the predesignated three divisions. The robustness of the model is such that, even with limited historic information, it can provide meaningful and dependable insurance premiums. The model has been extended to provide detailed regional vulnerability analysis. This should increase the societal financial resilience to the negative consequences of windstorm risks. | |
