Study On Forecasting Tourism Number In Guangdong Province Based On Time Series Model And Grey Model

Nowadays,with the improvement of the national economy and the change of people's ideas,more and more people regard going for a trip as a kind of leisure and entertainment.Although the huge flow of people brings huge economic benefits to the tourism market,it also causes pressure on the traffic and ecological environment of scenic spots.Therefore,it is particularly important to predict the flow of tourists in advance so as to take corresponding actions.Among many methods of forecasting passenger flow,time series analysis and grey model predication method are the most common two methods.A single model usually cannot inspect all the characteristics of the data,and may cause errors in the predication.Therefore,combining the characteristics of different models is an important subject in dealing with predication.This paper mainly focuses on the application research of time series model and grey model method in the prediction of tourists flow in Guangdong Province.The work of this paper can be summarized as follows:1)Three time series analysis and prediction models(including short-memory summation and autoregressive moving average(ARIMA)model and seasonal differential autoregressive moving average(SARIMA)model,and long-memory fractal autoregressive moving average(ARFIMA)model)were used to test the tourists number data set of Guangdong Province from 2001 to 2018.The test results show that ARFIMA model has better prediction effect in the long-term prediction.In the short-term prediction,ARIMA model has the worst prediction effect and deviates far from the real value.The prediction effect of SARIMA model is similar to that of ARFIMA model.2)Aiming at the long-memory ARFIMA model,this paper uses binomial and matrix knowledge to derive the specific fractional difference process,and tests it on the tourist number data set of Guangdong Province.The test results show that in the long-term prediction,the relative error between the real value and the predicted value of ARFIMA model remains in a reasonable range,and the prediction effect is better.3)Aiming at the forecasting effect of long-memory ARFIMA model,this paper combines the grey forecasting model(FGM(1,1).An improved FGM(1,1)-ARFIMA model based on ARFIMA model is proposed.The test results on the same data set show that in the long-term prediction,the improved FGM(1,1)-ARFIMA model can improve the forecasting accuracy of ARFIMA model to a certain extent.