پیش بینی تقاضای گردشگری خارجی (یک مطالعه موردی برای ایران)

Extended Abstract Forecasting is an essential analytical tool for tourism policy and planning. Business success, marketing decisions, government’s investment policy as well as macroeconomic policy are influenced by the accuracy of tourism forecasts, since the tourism product comprises a number of services that cannot be accumulated,  Accurate forecasts of tourism demand are paramount to ensure the availability of such services when demanded. on the other hand since univariate time series modeling has proved to be a very successful method for forecasting tourist arrivals,  In this article, along with study of ARFIMA and artificial intelligence methods ,we use the new hybrid approach combining ARFIMA and feed forward neural networks (FNN) proposed by  Aladagh et al. in ۲۰۱۲.For this purpose, ARFIMA, ANN and ARFIMA-ANN models are considered and compared their results in different time horizon using IRICA(The Islamic Republic of Iran Customs Administration) monthly time series of tourist arrivals to Iran from march ۱۹۹۹ to march ۲۰۱۲. According to MAPE, RMSE and MAE criteria, the forecasting performance of the ARFIMA-ANN model is better than the ARFIMA and ANN models at the horizons of ۶, ۱۲, ۱۸ and ۲۴ months and thus it can be used as suitable model for forecasting tourist arrivals to Iran.     Introduction Forecasting is an essential analytical tool for tourism policy and planning. Accurate forecasting of future tourist flow is essential to determine successful investments in the tourism industry for both the public and the private sectors. For the public sector, estimation of tourism demand is important in order to make efficient use of transportation and resources. For the private sector, such as airlines, good tourism forecasting is useful for planning aircraft, facilities and manpower needs (Chang and Liao, ۲۰۱۰:۲۱۵), despite of the importance of this issue, very little research has been done in this area in IRAN. Therefore In this article, ARFIMA, ANN and ARFIMA-ANN models were used for forecasting of country’s tourism time series.   Materials and Methods Since univariate time series modeling has proved to be a very successful method for forecasting tourist arrivals, it is also the method employed in this paper. In order to choose best model for forecasting tourist arrivals to Iran ARFIMA, ANN and ARFIMA-ANN methods were compared to forecast tourist flows to Iran at the horizons of ۶, ۱۲, ۱۸ and ۲۴ months using IRICA(The Islamic Republic of Iran Customs Administration) monthly time series of tourist arrivals to Iran from march ۱۹۹۹ to march ۲۰۱۲.   Discussion and Results According to MAPE, RMSE and MAE  criteria, the best results is obtained by hybrid(ARFIMA-ANN) method which means that ARFIMA-ANN thus it can be used as suitable model for forecasting tourist arrivals to Iran.   Conclusions The ARFIMA-ANN method is appropriate method to forecast foreign tourist to Iran and it will can be used as suitable model for forecasting tourist arrivals to Iran. References: Abrishami, H. and Mehrara, M. (۲۰۰۲). Applied econometrics (New approach), Tehran: University of Tehran Press. (In Persian) Aladag, C.H., Egrioglu, E. and Kadilar, C. (۲۰۱۲). Improvement in forecasting accuracy using the hybrid model of ARFIMA and feed forward neural network, American Journal of Intelligent Systems, ۲(۲):۱۲-۱۷. Archer, B.H. (۱۹۸۷). Demand forecasting and estimation, In Ritchie, J.R.B. and Goeldner, C.R. (Eds) Travel, tourism and hospitality research, New York: Wiley. Aslanargun, A., Mammadov, M., Yazici, B. and Yolacan, S. (۲۰۰۷). Comparison of ARIMA, neural networks and hybrid models in time series: Tourist arrival forecasting, Journal of Statistical Computation and Simulation, ۷۷: ۲۹–۵۳. Athanasopoulos, G., Hyndman, R.J., Song, H. and Wu, D.C. (۲۰۱۱). The tourism forecasting competition, International Journal ofForecasting, ۲۷: ۸۲۲–۸۴۴. Burger, C.J.S.C., Dohnal, M., Kathrada, M. and Law, R. (۲۰۰۱). A practitioners guide to time-series methods for tourism demand forecasting–A case study of Durban, South Africa, Tourism Management, ۲۲: ۴۰۳–۴۰۹. Claveria, O., Monte, E. and Torra, S. (۲۰۱۴). Tourism demand forecasting with neural networkmodels: Different ways of treating information, International Journal of Tourism Research, ۱-۲۰, DOI: ۱۰.۱۰۰۲/jtr.۲۰۱۶. Claveria, O. and Torra, S. (۲۰۱۴). Forecasting tourism demand to Catalonia: Neural networks vs. time series models, Economic Modelling, ۳۶: ۲۲۰-۲۲۸. Chang, Y.W. and Liao, M.Y. (۲۰۱۰). A seasonal ARIMA model of tourism forecasting: The case of Taiwan, Asia Pacific journal of Tourism research, ۱۵(۲): ۲۱۵-۲۲۱. ۱۰. Chu, F.L. (۲۰۰۸). A fractionally integrated autoregressive moving average approach to forecasting tourism demand, Tourism Management, ۲۹: ۷۹-۸۸. ۱۱. Chu, F.L. (۲۰۰۹). Forecasting tourism demand with ARMA-based methods, Tourism Management, ۳۰:۷۴۱-۷۵۱. ۱۲. Cuhadar, M. (۲۰۱۴). Modelling and forecasting inbound tourism demand to Istanbul–A comparative analysis, European Journal of Business and Social Sciences, ۲:۱۲. ۱۳. Doornik, A. and Ooms, M. (۱۹۹۹). A package for estimating, forecasting and simulating ARIFMA models: Help of ARIFMA package ۱.۰ for Ox, from fmwww.bc.edu/ec-p/software/ox/Ox.arfima.v۲.۱.pdf. ۱۴. Gil-Alana, L.A. (۲۰۰۵). Modeling international monthly arrivals using seasonal univariate long-memory processes, Tourism Management, ۲۶: ۸۶۷-۸۷۸. ۱۵. Granger C.W.J and Joyeux, R. (۱۹۸۰). An introduction to long-memory time series models and fractional differencing- J, Time Series Anal, ۱(۱):۱۵-۲۹. ۱۶. Hwang, K.P. and Day, Y.J. (۲۰۱۳). Tourism revenue forcasting: A hybrid model approach, Actual Problems of Economics, ۱۴۱(۳): ۴۷۸-۴۸۴. ۱۷. Kuan-Yu, C. (۲۰۱۱). Combining linear and nonlinear model in forecasting tourism demand, Expert Systems with Applications, ۳۸: ۱۰۳۶۸–۱۰۳۷۶. ۱۸. Lewis, C.D. (۱۹۸۲). International and business forecasting methods, London: Butterworths. ۱۹. Lildhold, P. (۲۰۰۰). Long memory an ARFIMA modeling, University of Aarhus. ۲۰. Loganathan, N. and Ibrahim, Y. (۲۰۱۰). Forecasting international tourism demand in Malaysia using Box Jenkins sarima application, South Asian Journal of Tourism and Heritage, ۳(۲): ۵۰-۵۹. ۲۱. Saayman, A. and Saayman, M. (۲۰۱۰). Forecasting tourist arrivals in South Africa, Acta Commercii, ۱۰: ۲۸۱-۲۹۳. ۲۲. Shitan, M. (۲۰۰۸). Time series modelling of tourist arrivals to Malaysia, InterStat, (October): ۱-۱۲. ۲۳. Song, H. and Li, G. (۲۰۰۸). Tourism demand modelling and forecasting–A review of recent research, Tourism Management, ۲۹ (۲): ۲۰۳-۲۲۰. ۲۴. Song, H. and Witt, S.F. (۲۰۰۶). Forecasting international tourist flows to Macau, Tourism Management, ۲۷: ۲۱۴-۲۲۴. ۲۵. Teixeira, J. P. and Fernandes, P.O. (۲۰۱۲). Tourism time series forecast-different ANN architectures with time index input, Procedia Technology, ۵: ۴۴۵-۴۵۴. ۲۶. Tseng, F.M., Yu, H.C. and Tzeng, G.H. (۲۰۰۲). Combining neural network with seasonal time series ARIMA model, Technological Forecasting and Social Change, ۶۹(۱):۷۱–۸۷. ۲۷. UNWTO. (۲۰۱۲). ‘Why tourism?’ [Online article]. Available from: http: // www۲.unwto.org/en/ content/why- tourism, [Accessed ۱۵ September ۲۰۱۲]. ۲۸. Xiu, J. and Jin, Y. (۲۰۰۷). Empirical study of ARFIMA model based on fractional differencing, Physica A, ۳۷۷:۱۳۸ –۱۵۴. ۲۹. Zhang, G.P. (۲۰۰۳). Time series forecasting using a hybrid ARIMA and neural network model, Neurocomputing, ۵۰: ۱۵۹-۱۷۵. ۳۰. Zhang, G.P., Patuwo, E.B. and Hu, M.Y. (۱۹۹۸). Forecasting with artificial neural networks: the state of the art, International Journal of Forecasting, ۱۴: ۳۵–۶۲. ۳۱. Zhang, G.P., Patuwo, E.B. and Hu M.Y. (۲۰۰۱). A simulation study of artificial neural networks for nonlineartime-series forecasting, Computer & Operation Research, ۲۸: ۳۸۱–۳۹۶.