Objective To establish a model to predict the trend of varicella incidence in Baoshan District,Shanghai in 2024,and to provide a scientific basis for early warning,effective prevention and control of varicella.Methods Utilizing the data of reported varicella cases in Baoshan District from 2010 to 2022,a seasonal autoregressive integrated moving average model was constructed using the R programming language.Validate the model with data from 2023,and finally use the validated model to predict the trend of chickenpox epidemic in 2024.Results The average annual incidence rate of varicella in Baoshan District from 2010 to 2022 was 69.67/105.The number and incidence of cases both decreased from 2020 onwards.The time series of monthly varicella incidence data in Baoshan District was stable,and had obvious seasonality.ARIMA(1,0,1)(0,1,1)12was the best-fitted model.Ljung-Box test for the model residuals showed white noise.The actual values in January to December 2023 were all within the 95％confidence intervals of the predicted values.The prediction showed that the number of varicella cases in Baoshan District in 2024 was 762,and there would be two peaks of incidence throughout the year.Conclusions ARIMA(1,0,1)(0,1,1)12 model can fit the trend of varicella incidence in Baoshan District better,and can be used to predict the short-term incidence of varicella in Baoshan District.
关键词
季节性自回归差分滑动平均模型模型/水痘/趋势/预测/R语言
Key words
SARIMA model/varicella/trend/forecast/R programming language
维普
万方数据