Abstract: This paper selects the monthly data of the number of visitors to Shanghai from January 2004 to August 2012. Based on the general time series prediction method, intervention analysis is introduced, and the R software is used to predict the time series. First of all, through analysis, the trend effect and seasonal effect of the series are found, so a product season model is used to fit the time series and predict the the number of visitors to Shanghai for the next 8 months. Secondly, after pre-processing the original data and determining the time point of the Expo impact, the time series is divided into two parts, by the time point then the method of intervention analysis is used to establish an intervention combination model to predict the number of visitors to Shanghai for the next 8 months. Finally, the prediction results of the two models are compared by calculating their relative errors. By comparison, it is found that the prediction effect of the intervention model is better, indicating that in the presence of emergencies or major policies, it is better to use an intervention model to analyze and predict the time series.

Key words: Forecast ,  Inbound tourist ,  Time series ,  , Intervention model