Predicting hurricane wind damage by claim payout based on Hurricane Ike in Texas

The increasing occurrence of natural disasters and their related damage have led to a growing demand for models that predict financial loss. Although considerable research on the financial losses related to natural disasters has found significant predictors, there has been a lack of comprehensive study that addresses the relationship among vulnerabilities, natural disasters, and the economic losses of individual buildings. This study identifies the vulnerability indicators for hurricanes to establish a metric to predict the related financial loss. We classify hurricane-prone areas by highlighting the spatial distribution of losses and vulnerabilities. This study used a Geographical Information System (GIS) to combine and produce spatial data and a multiple regression method to establish a wind damage prediction model. As the dependent variable, we used the value of the Texas Windstorm Insurance Association (TWIA) claim payout divided by the appraised values of the buildings to predict real economic loss. As independent variables, we selected a hurricane indicator and built environment vulnerability indicators. The model we developed can be used by government agencies and insurance companies to predict hurricane wind damage.