Sweet Potato Yield Prediction for Index Insurance in North Carolina

Abstract

Agriculture index insurance is an innovative topic that has not been well studied in the United States. North Carolina produces 1.7 billion pounds of sweet potatoes in 2020, but currently, there is no insurance to reduce the financial risk of farmers. As a result, index insurance focusing on North Carolina sweet potato farmers can be profitable. In this study, the precipitation is forecasted by the linear model using the first lag and seasonal factors. The predicted precipitations from May to September are then used to predict the yield. The precipitation model has significant factors for Season3, which represents July to September, the rainy season of North Carolina; the yield model has a significant variable of September, which is the harvest season of sweet potatoes in North Carolina. The precipitation model falls short of predicting the exact value of precipitation, but it catches the trend and seasonality. Despite the insensitivity of the precipitation model, the yield is predicted relatively accurate. The result of this study can be used to design the thresholds of the index insurance. Insurance companies can use thresholds to design insurance plans with different premiums.