Comparing the results of two models in prediction of dysentery incidence
Jia Lei · Zhongguo redai yixue · 2013
Objective To ccompare the differences between winters multiplication model and autoregressive integrated moving average model(ARIMA) in predicting the incidence of dysentery in Beijing.Methods The monthly incidence data of dysentery from January 2007 to December 2012 in Beijing were collected and modeling the data with winters multiplication model and ARIMA.The results of predictingthe incidence of dysentery in the first quarter of 2013 in Beijing were evaluated.Results After the assessment of fit of these two models using data in 2012,measured by prediction percentage error,winters multiplication mode(l 1.13%)was found to be better than ARIMA(6.80%).The predicting incidence rates of dysentery by using the winters multiplication model in the first quarter of 2013 were 1.82/100000,1.54/100000 and 1.85/100000.Conclusions winters multiplication model could well reflect the trend of the incidence of dysentery in Beijing and it was suitable for predict ing the future trend dysentery.