Abstract: The effect of two mathematical models on southern rice black-streaked dwarf virus was studied to provide reference for the prediction of the disease. Based on the original data of Incidence of late southern rice black-streaked dwarf virus in Huazhou City, Guangdong Province, Polynomial regression method and Grey system GM (1,1) prediction method are applied to build the prediction model of southern rice black-streaked dwarf virus incidence rate and to analyze the prediction results. The results show that, the late southern rice black-streaked dwarf virus incidence prediction models established by polynomial regression and grey system GM (1, 1) are: y=5.2530-5.8923x 2.1280x2-0.1559x3 and y(1)(k 1)=-51.1797e-0.4798k 66.8097. The two prediction models can well predict the incidence trend of southern rice black-streaked dwarf virus, and provide an accurate and effective method for the prediction of the disease. The predictive model can be applied to the project of disease prevention and mitigation.