Malezya'ya Gelen Turist Sayılarını Tahmin Etmek için Hibrit SARIMA-ANN Yaklaşımı

Abstract

Turist sayılarının doğru tahmin edilmesi, stratejik planlama ve ekonomik büyüme için çok önemlidir, özellikle de turizmin Gayri Safi Yurt İçi Hasıla'ya önemli katkı sağlayan Malezya'da. Mevsimsel Otoregresif Entegre Hareketli Ortalama (SARIMA) gibi geleneksel tahmin modelleri, doğrusal ve mevsimsel kalıpları etkili bir şekilde yakalar, ancak doğrusal olmayan bağımlılıklarla başa çıkmakta zorlanır. Tersine, Yapay Sinir Ağı (ANN) doğrusal olmayan ilişkileri modellemede mükemmeldir, ancak zamansal bağımlılıkları hesaba katmakta başarısız olabilir. Bu sınırlamaları gidermek için, bu çalışma tahmin doğruluğunu artırmak için her iki modelin güçlü yönlerini birleştiren hibrit bir SARIMA-ANN modeli önermektedir. Bu çalışmada, önce SARIMA uygulanarak doğrusal ve mevsimsel bileşenler çıkarılmış, ardından kalan doğrusal olmayan kalıpları yakalamak için ANN kullanılarak kalıntıları modellenmiştir. Ampirik sonuçlar, SARIMA-ANN modelinin zaman serisi verilerindeki doğrusal ve doğrusal olmayan kalıpları etkili bir şekilde yakalama yeteneği sayesinde yüksek doğrulukta bir hibrit model sağlayabildiğini göstermektedir. SARIMA mevsimselliği ve uzun vadeli eğilimleri etkili bir şekilde yakalarken, ANN kalıntılar içindeki karmaşık doğrusal olmayan ilişkileri belirleyerek sınırlamalarını aşmaktadır. Bu birleşik model, her iki modelin güçlü yönlerini kullanarak tahmin hatalarını azaltır ve her bir modelin tek başına aşamadığı sınırlamaları giderir. Bu bulgular, turist gelişlerini tahmin etmek için hibrit zaman serisi modellemesinin etkinliğini vurgulamakta ve mevsimsel ve doğrusal olmayan modellerin mevcut olduğu diğer alanlarda da potansiyel olarak uygulanabileceğini göstermektedir. Bu çalışma, ekonomik göstergeler, sosyal medya trendleri veya turizm talebini etkileyen diğer dış faktörler gibi dış değişkenleri içermediğini ima etmektedir, bu da modelin ani değişiklikleri yakalama yeteneğini sınırlayabilir.

Keywords

• hybrid sarima - ann
• tourist arrivals
• time series forecasting
• hybrid model

Hybrid SARIMA-ANN approach for forecasting tourist arrivals in Malaysia

Abstract

Accurate forecasting of tourist arrivals is essential for strategic planning and economic growth, particularly in Malaysia, where tourism is a key contributor to Gross Domestic Product. Traditional forecasting models, such as the Seasonal Autoregressive Integrated Moving Average (SARIMA), effectively capture linear and seasonal patterns but struggle with nonlinear dependencies. Conversely, Artificial Neural Network (ANN) excels at modeling nonlinear relationships but may fail to account for temporal dependencies. To address these limitations, this study proposes a hybrid SARIMA-ANN model that integrates the strengths of both models to improve forecasting accuracy. In this study, a hybrid SARIMA-ANN model is developed by first applying SARIMA to extract linear and seasonal components, followed by modeling its residuals using an ANN to capture any remaining nonlinear patterns. Empirical results demonstrate that the SARIMA-ANN model can provide a hybrid model with high accuracy due to its ability to effectively capture both linear and nonlinear patterns in time series data. SARIMA effectively captures seasonality and long-term trends, while ANN overcomes its limitations by identifying complex nonlinear relationships within the residuals. This combined model reduces forecasting errors by leveraging the strengths of both models, addressing limitations that each model individually cannot overcome. These findings highlight the effectiveness of hybrid time series modeling for forecasting tourist arrivals, suggesting its potential application in other domains where seasonal and nonlinear patterns exist. This study implies that it does not include external variables such as economic indicators, social media trends, or other external factors that influence tourism demand, which may limit the model’s ability to capture sudden changes.

Keywords

• hybrid sarima-ann
• time series forecasting
• tourist arrivals
• hybrid model

References

1. Abellana, D.P.M., Rivero, D.M.C., Aparente, M.E., & Rivero, A. (2021). Hybrid SVR-SARIMA model for tourism forecasting using PROMETHEE II as a selection methodology: a Philippine scenario. Journal of Tourism Futures, 7(1), 78-97.
2. Azad, A.S., Sokkalingam, R., Daud, H., Adhikary, S. K., Khurshid, H., Mazlan, S. N. A., & Rabbani, M. B. A. (2022). Water level prediction through hybrid SARIMA and ANN models based on time series analysis: Red hills reservoir case study. Sustainability, 14(3), 1843.
3. Bouhaddour, S., Sbihi, M., Guerouate, F., Saadi, C. (2024). Assessing tourism prediction models: A comparative study of SARIMA, Random Forest, and LSTM, Considering nonlinear trends and the influence of COVID-19. International Information and Engineering Technology Association, 29(6), 2443-2454.
4. Box, G.E.P., Jenkins, G.M., Reinsel, G.C. (2008). Time series analysis: Forecasting and control (4th ed.). John Wiley and Sons, Hoboken, New Jersey.
5. Chikobvu, D., Makoni, T. (2019). Statistical modelling of Zimbabwe’s international tourist arrivals using both symmetric and asymmetric volatility models. Journal of Economic and Financial Sciences, 12(1), 1-10.
6. Darban, Z.Z., Webb, G.I., Pan, S., Aggarwal, C.C., Salehi, M. (2022). Deep learning for time series anomaly detection: A survey. ACM Computing Surveys, 57(1), 1-42.
7. Farsi, M., Hosahalli, D., Manjunatha, B.R., Gad, I., Atlam, E.S., Ahmed, A., Elmarhomy, G., Elmarhoumy, M., Ghoneim, O.A. (2021). Parallel genetic algorithms for optimizing the SARIMA model for better forecasting of the NCDC weather data. Alexandria Engineering Journal, 60(1), 1299-1316.
8. Hadwan, M., Al-Maqaleh, B.M., Al-Badani, F.N., Khan, R.U., Al- Hagery, M.A. (2022). A hybrid neural network and box-jenkins models for time series forecasting. Computers, Materials and Continua, 70(3), 4829-4845.

9. Hewamalage, H., Bergmeir, C., Bandara, K. (2021). Recurrent neural networks for time series forecasting: Current status and future directions. International Journal of Forecasting, 37(1), 388- 427.
10. Huang, J., Zhang, C. (2024). Daily tourism demand forecasting with the iTransformer Model. Sustainability, 16(23), 10678.
11. Hyndman, R.J., Athanasopoulos, G. (2018). Forecasting: Principles and practice. OTexts, Melbourne, Australia.
12. Joseph, V.R. (2022). Optimal ratio for data splitting. Statistical Analysis and Data Mining, 15(4), 531-538.
13. Justin, S., Waoo, A., Waoo, A.A. (2024). Performance analysis of activation functions in neural networks. ShodhKosh: Journal of Visual and Performing Arts, 5(1), 588-598.
14. Kuiper, R.M. (2021). AIC-type Theory-Based Model selection for structural equation models. Structural Equation Modeling: A Multidisciplinary Journal, 29(1), 151-158.
15. Mapuwei, T.W., Bodhlyera, O., Mwambi, H. (2020). Univariate time series analysis of short-term forecasting horizons using artificial neural networks: The case of public ambulance emergency preparedness. Journal of Applied Mathematics, 2020(1), 2408698.
16. Misengo, E.E., Prastyo, D. D., and Kuswanto, H. (2023). Modeling and Forecasting Monthly Tourist Arrivals to the United States and Indonesia Using ARIMA Hybrids of Multilayer Perceptron Models. AIP Conference Proceedings, 2540.
17. Mukaram, M. Z., and Yusof, F. (2017). Solar radiation forecast using hybrid SARIMA and ANN model. Malaysian Journal of Fundamental and Applied Sciences, 13, 346–350.
18. My, H., Setyowati, S.L., Notodiputro K.A., Angraini, Y., Mualifah, L.N.A. (2024). Comparison of seasonal ARIMA and support vector machine forecasting method for international arrival in Lombok. Jambura Journal of Mathematics, 6(2), 212-219.
19. Parviz, L. (2020). Comparative evaluation of hybrid sarima and machine learning techniques based on time-varying and decomposition of precipitation time series. Journal of Agricultural Science and Technology, 22(2), 563-578.
20. Pushpa, P., Kumar J, Vikram, G. (2022). Hybrid Sarima-Ann model for forecasting monthly wholesale price and arrival series of tomato crop. Biological Forum – An International Journal, 14(4a), 591- 596.
21. Rajalakshmi, V., Vaidyanathan, S.G. (2022). Hybrid time-series forecasting Models for traffic flow prediction. Promet - Traffic - Traffico, 34(4), 537-549.
22. Saba, T., Shahzad, M. N., Iqbal, S., Rehman, A., & Abunadi, I. (2022). A New Hybrid SARFIMA-ANN Model for Tourism Forecasting. Computers, Materials & Continua, 71(3), 4785–4801.
23. Wu, D. C. W., Ji, L., He, K., and Tso, K. F. G. (2021). Forecasting Tourist Daily Arrivals With A Hybrid Sarima–Lstm Approach. Journal of Hospitality and Tourism Research, 45(1), 52–67.
24. Yollanda M, Devianto D. (2020). Hybrid Model of Seasonal ARIMA-ANN to forecast tourist arrivals through Minangkabau International Airport. Proceedings of the 1st International Conference on Statistics and Analytics, ICSA 2019, 2-3 August 2019, Bogor, Indonesia.
25. Yu Y, Wang Y, Gao S, Tang Z. (2017). Statistical modeling and prediction for tourism economy using dendritic neural network. Computational Intelligence and Neuroscience, 2017(1), 7436948.