ARMA model for predicting the number of new outbreaks of newcastle disease during the month

After the outbreak of avian influenza and swine influenza, people realize that it is very important to surveil the infectious diseases outbreak of livestock and poultry in flocks. Owing to specificity of animal husbandry production, it is so difficult to control the epidemic of animal infectious diseases using the medical treatment only. So it is necessary to construct the mathematical model for predicting the outbreak and development of animal infectious diseases, which can make the loss to the minimum. Time series models constitutes one of the most important applications in the animal disease prevention and control system. Specific prediction of animal disease outbreaks would be helpful to local animal health departments for appropriate preparedness and to take selective preventive measures in areas at risk of epidemics. ARMA model is considered as one of the most accurate time series models and is applied extensively. In this paper, an attempt is made to show how ARMA model is applied to predicting the number of new outbreaks of ND during the month in a province in china, and to establish some corresponding mathematical predicting models. Finally, ARMA(0,1) was chosen as predicting model and the parameters were estimated using SPSS. After parameters estimation, MA1 parameter and constant parameter were obtained as, 0.682 and -0.015 respectively. MAE for prediction model is equal to 1.257, and it was concluded that ARMA(0,1) provided reliable and accurate forecasts of ND outbreaks.