Abstract: OBJECTIVE The ARIMA model was used to predict the prevalence rate of nosocomial infection and to provide decision making basis for the prevention and control of nosocomial infection in the whole hospital. METHODS A retrospective survey was conducted to collect the prevalence rate of nosocomial infection in a hospital from Jan. 2012 to Dec. 2017(a total of 72 months).A model was constructed by ARIMA time series model, and the fitting and prediction effects of the model were evaluated. RESULTS After comparing the alternative models,it was confirmed that the fitting effect of ARIMA (1,0,0) model was relatively good. The actual prevalence rates in the first six months of 2018 were within 95% confidence interval of the predicted values. CONCLUSION The model can accurately predict the prevalence rate of nosocomial infection,and is helpful to formulate and implement the preventive and control measures related to nosocomial infection management.