Insurance claims forecasting for extreme weather events that result in large scale destruction such as hurricanes, wildfires, floods, etc. is an important planning activity for insurance firms and any process improvements that can enhance the accuracy and quality of the forecasts should be welcome. This article provides an introduction to the forecasting methodology for insurance claims payouts and then discuss potential use of machine learning techniques to enhance the forecasting process.
Prediction is difficult, especially about the future!