Comparison of Forecasting Performance of ARIMA LSTM and HYBRID Models for The Sales Volume Budget of a Manufacturing Enterprise

Abstract

This study aims to create a monthly sales quantity budget by making use of the previous income data of an enterprise operating within the construction sector, which is considered the locomotive of the economy. For estimating time-series of sales as a linear model ARIMA (Auto-Regressive Integrated Moving Average), as nonlinear model LSTM (Long Short-Term Memory) and a HYBRID (LSTM and ARIMA) model built to improve system performance compared to a single model was used. As a result of the study, Mean Square Error (MSE), Root Mean Square Error (RMSE), Mean Absolute Percentage Error (MAPE) and Mean Absolute Error (MAE) values obtained from each of the methods used in the application were compared, and a monthly sales volume budget was created for 2017 with all the methods used. When the MAPE and MSE values obtained from each of these methods were compared, the best performance was the Hybrid model that gave the lowest error, and in addition, the fact that all of the application models got very realistic results by using the historical data showed the success of the predictions.

Keywords

• Sales Budget
• Time Series Forecast
• Hybrid Model
• ARIMA
• LSTM

References

1. Açıcı, K., Aşuroğlu, T., & Oğul, H. (2017). Information retrieval in metal music sub-genres. 2017 25th Signal Processing and Communications Applications Conference (SIU). Antalya, TURKEY: IEEE.
2. Adeva, J. J., Beresi, U. C., & Calvo, R. A. (2005). Accuracy and Diversity in Ensembles of Text Categorisers. CLEI Electronic Journal, 8(2), 1-12. doi:DOI: 10.19153/cleiej.12.1.1
3. Ahmetoğlu, M. E. (2018). Veri Merkezi Sektöründe Satış Tahminleme Modeli ve Özel Kurum Üzerinde Ugulaması. İstanbul: Bahçeşehir Üniversitesi.
4. Aladağ, Ç. H., Eğrioğlu, E., & Kadılar, C. (2009). Forecasting Nonlinear Time Series with a Hybrid Methodology. Applied Mathematics Letters(22), 1467-1470. doi:https://doi.org/10.1016/j.aml.2009.02.006
5. Albayrak, A. S. (2010). ARIMA Forecasting of Primary Energy Production and Consumption in Turkey: 1923–2006. Enerji, Piyasa ve Düzenleme, 24-50.
6. Armano, G., Marchesi, M., & Murru, A. (2005). A Hybrid Genetic-Neural Architecture for Stock Indexes Forecasting. Information Sciences(170), 3-33.
7. Ateşonğun, A. A. (2015). ARIMA ve Yapay Sinir Ağları (YSA) Kullanılarak Hibrit Tahmin Modeli Geliştirilmesi. Ankara: Başkent Üniversitesi Fen Bilimleri Enstitüsü.
8. Atienza, R. (Posted on 2017, 4 26). LSTM by Example using Tensorflow (Text Generate). towardsdatascience.com: https://towardsdatascience.com/lstm-by-example-using-tensorflow-feb0c1968537

9. Babu, C. N., & Reddy, B. E. (2014). A Moving-Average Filter Based Hybrid ARIMA–ANN Model for Forecasting Time Series Data. Applied Soft Computing(23), 27-38. doi:https://doi.org/10.1016/j.asoc.2014.05.028
10. Bengio, Y., Patrice, S., & Frasconi, P. (1994). Learning long-term dependencies with gradient descent is difficult. IEEE transaction on neural networks, 157-166.
11. Bengio, Y., Simard, P., & Franscon, P. (2015). Learning Long-Term Dependencies with Gradient Descent is Diffucult. IEEE Trans. Neural Networks, 5(2), 157-166.
12. Bhardwaj, S., M.Gadre, V., & Chandrasekhar, E. (2020). Statistical Analysis of DWT Coefficients of fGn Processes Using ARFIMA (p,d,q) Models. Physica A: Statistical Mechanics and its Applications, 547, 124404. doi:10.1016/j.physa.2020.124404
13. Bircan, H., & Karagöz, Y. (2003). Box-Jenkins Modelleri ile Aylık Döviz Kuru Tahmini Üzerine Bir Uygulama. Kocaeli Üniversitesi Sosyal Bilimler Enstitüsü Dergisi, 6(2), 49-62. Şubat 10, 2017 tarihinde http://dergipark.gov.tr/download/article-file/252067
14. Box, P., & Jenkins, G. (1976). Time Series Analysis:Forecasting and Control. Holden-day Inc.
15. Brownlee, J. (2020). Machine Learning Mastery. https://machinelearningmastery.com/train-test-split-for-evaluating-machine-learning-algorithms/, Jully, 2020.
16. Bukhari, A. H., Raja, M. A., Sulaiman, M., Islam, S., Shoaib, M., & Kumam, P. (2020). Fractional Neuro-Sequential ARFIMA-LSTM for Financial Market Forecasting. IEEE ACCESS, 8, 71326-71338. doi:10.1109/ACCESS.2020.2985763
17. Büyükşahin, Ü., & Ertekin, Ş. (2018). Ampirik Kip Ayrışımı ve Hibrit Metot Kullanarak Zaman Serisi Tahmini. Sinyal İşleme ve İletişim Uygulamaları (SİU) Kurultayı. IEEE.
18. Chollet, F. a. (2015). Keras. chollet2015keras. \url{https://keras.io, 2020.04.12
19. Contreras, J., Espinola, R., Nogales, F., & Conejo, A. (2003). ARIMA models to predict next-day electricity prices. IEEE Tans. Power Syst., 1014-1020.
20. Düzyol, M. (1997). Türkiye’de Bina İnşaatı Sektörü ve 1990-2010 Dönemi Bölgesel İhtiyaç Tahmini. Ankara: DPT: 2484.
21. Ediger, V. Ş., & Akar, S. (2007). ARIMA Forecasting of Primary Energy Demand by Fuel in Turkey. Energy Policy, 1701-1708. doi:https://doi.org/10.1016/j.enpol.2006.05.009
22. Erdoğdu, E. (2007). Electricity Demand Analysis Using Cointegration and ARIMA Modelling: A Case Study of Turkey. Energy Policy, 1129-1146. doi:https://doi.org/10.1016/j.enpol.2006.02.013
23. Fattah, J., Ezzine, L., Aman, Z., Moussami, H. E., & Lachhab, A. (2018). Forecasting of Demand Using ARIMA Model. International Journal of Engineering Business Management, 10, 1-9. doi:10.1177/1847979018808673
24. Greenwood, J., & Hercowitz, Z. (1991). The Allocation of Capital and The Time Over The Business Cycle. Journal of Political Economy(99), 1188-1214. doi:http://dx.doi.org/10.1086/261797
25. He, G., & Deng, Q. (2012). A Hybrid ARIMA and Neural Network Model to Forecast Particulate. Matter Concentration in Changsha. https://www.isiaq.org/docs/PDF%20Docs%20for%20Proceedings/1F.3.pdf
26. Hocaoğlu, F. O., Kaysal, K., & Kaysal, A. (2015). Yük Tahmini İçin Hibrit (YSA ve Regresyon) Model. Akademik Platform, 33-39. doi:DOI: 10.5505/apjes.2015.94695
27. Hochreiter S. (1998). The vanishing gradient problem during learning recurrent neural nets and problem solutions. International Journal of Uncertainty, Fuzziness and Knowledge-Based System, 107-116.
28. Hochreiter, S., & Schmidhuber, J. (1997). Long Sort Term Memory. Neural Computation, 1735-1780.
29. Ioannou, K., Birbilis, D., & Lefakis, P. (2011). A Method for Predicting the Possibility of Ring Shake Appearance on Standing Chestnut Trees. Journal of Environmental Protection and Ecology, 295-304. https://docs.google.com/a/jepe-journal.info/viewer?a=v&pid=sites&srcid=amVwZS1qb3VybmFsLmluZm98amVwZS1qb3VybmFsfGd4OjJkYzQwODIyZjE4ZmJmMzQ
30. Kang, E. (2018). Github: https://github.com/llSourcell/LSTM_Networks/blob/master/LSTM%20Demo.ipynb
31. Khandelwal, I., Adhikari, R., & Verma, G. (2015). Time Series Forecasting using Hybrid ARIMA and ANN Models based on DWT Decomposition. Procedia Computer Science(48), 173-179.
32. Khashei M., H. S. (2008). A New Hybrid Artificial Neural Networks and Fuzzy Regression Model for Time Series Forecasting. Fuzzy Sets and Systems(159), 769-786. doi:https://doi.org/10.1016/j.fss.2007.10.011
33. Khashei, M., & Bijari, M. (2011). A Novel Hybridization of Artificial Neural Networks and ARIMA Models for Time Series Forecasting. Applied Soft Computing(11), 2664-2675.
34. Khashei, M., Hejazi, S. R., & Bijari, M. (2008). A New Hybrid Artificial Neural Networks and Fuzzy Regression Model for Time Series Forecasting. Fuzzy Sets and Systems(159), 769-786. doi:https://doi.org/10.1016/j.fss.2007.10.011
35. Kılıç, R., & Demirbaş, E. (2012). Türkiye’de Kamu İnşaat Harcamalarının Belirleyicileri ile Ekonomik Büyüme Arasındaki İlişki. Akademik Yaklaşımlar Dergisi, 3(2), 84-97.
36. Kim, H., & Shin, K. (2007). A Hybrid Approach Based on Neural Networks and Genetic Algorithms for Detecting Temporal Patterns in Stock Markets. Applied Soft Computing(7), 569-576.
37. Koutroumanidis, T., Ioannou, K., & Zafeiriou, E. (2011). Forecasting Bank Stock Market Prices with A Hybrid Method: The Case of Alpha Bank. Journal of Business Economics and Management, 12(1), 144-163. doi:https://doi.org/10.3846/16111699.2011.555388
38. Koutroumanidis, T., Ioannoub, K., & Arabatzis, G. (2009). Predicting Fuelwood Prices in Greece With the Use of ARIMA Models, Artificial Neural Networks and a Hybrid ARIMA–ANN Model. Energy Policy(37), 3627-3634. doi:https://doi.org/10.1016/j.enpol.2009.04.024
39. Lin, T., Guo, T., & Aberer, K. (2017). Hybrid Neural Networks for Learning the Trend in Time Series. Proceedings of the Twenty-Sixth International Joint Conference on Artificial Intelligence, (s. 2273-2279). Melbourne, Australia. doi:https://doi.org/10.24963/ijcai.2017/316
40. Luxhoj, J., Riis, J., & Stensballe, B. (1996). A Hybrid Econometric-Neural Network Modeling Approach for Sales Forecasting. International Journal of Production Economics(45), 175-192.
41. Namın, S. S., & Namın, A. S. (2018). Forecasting Economic and Financial Time Series: ARIMA vs. LSTM. Lubbock, TX, USA: Texas Tech University. doi:https://arxiv.org/ftp/arxiv/papers/1803/1803.06386.pdf
42. Newbold, P. (1983). ARIMA Model Building and The Time Series Analysis Approach to Forecasting. Journal of Forecasting, 2(1), 23-35. doi: https://doi.org/10.1002/for.3980020104
43. Olah, C. (2015, August 27). Understanging LSTM Networks. Nisan 10, 2019 tarihinde colah.github.io: colah.github.io/posts/2015-08-Understanding-LSTMs/.
44. Oliveira, M., & Torgo, L. (2014). Ensembles for Time Series Forecasting. In Proceedings of the Sixth Asian Conference on Machine Learning, (s. 360-370). http://proceedings.mlr.press/v39/oliveira14.pdf adresinden alındı
45. Opitz, D., & Maclin, R. (1999). Popular Ensemble Methods: An Empirical Study. Journal of Artificial Intelligence Research, 11, 169–198. doi:https://doi.org/10.1613/jair.614
46. Pablo, B. J., Hilda, C., Xavier, A., Diego, J. J., Felipe, S., & Henry, B. (2016). Artificial Neural Network and Monte Carlo Forecasting with Generation of L-Scenarios. 2016 Intl IEEE Conferences on Ubiquitous Intelligence & Computing, Advanced and Trusted Computing, Scalable Computing ond Communications, Cloud and Big Data Computing, Internet of People and Smart World Congress, (s. 665-670). doi:10.1109/UIC-ATC-ScalCom-CBDCom-IoP-SmartWorld.2016.0110
47. Pagourtzi, E., Assimakolopulos, V., Hatzichristons, T., & French, N. (2003). Real Estate Appraisal: A Review of Valuation Methods. Journal of Property Investment & Finance, 21(4), 383-401.
48. Papagera, A., Ioannou, K., Zaimes, G., Iakovoglou, V., & Simeonidou, M. (2014). Simulation and Prediction of Water Allocation Using Artificial Neural Networks and a Spatially Distributed Hydrological Model. Agris on-line Papers in Economics and Informatics, 6(4), 101-111. https://ageconsearch.umn.edu/record/196580/.
49. Pascanu, R., Mikolov, T., & Bengio, Y. (2013). On the difficult of training recurrent neural networks. ICML, 1310-1318.
50. Prieto, M. P., Uribbarri, P. M., & Tardioli, G. (2019). Applying Modeling and Optimization Tools to Existing City Quarters. Urban Energy Systems for Low-Carbon Cities, 333-414. doi:10.1016/B978-0-12-811553-4.00010-X
51. Rossum, G. V. (2001). Python. (T. M. Python) https://docs.python.org/3/2019.05.17
52. Sallehuddin, R., Shamsuddin, S. M., Hashim, S. Z., & Abraham, A. (2007). Forecasting Time Series Data Using Hybrid Grey Relational Artificial Neural Network and Auto Regressive Integrated Moving Average Model. ICS AS CR 2007, (s. 573-605). https://www.researchgate.net/publication/287522177_Forecasting_time_series_data_using_hybrid_grey_relational_artificial_neural_network_and_auto_regressive_integrated_moving_average_model.
53. Sarı, M. (2016, Ocak). Yapay Sinir Ağları ve Bir Otomotiv Firmasında Satış Talep Tahmini Uygulaması. Yüksek Lisans Tezi. Sakarya, Türkiye: Sakarya University.
54. Smith, M. J. (2018). Statistical Analysis (2 b.). Edinburgh: The Winchelsea Press, Drumlin Security Ltd. https://www.statsref.com/StatsRefSample.pdf
55. Soy Temür, A., Akgün, M., & Temür, G. (2019). Predicting Housing Sales In Turkey Using Arima, Lstm And Hybrid Models. Journal of Business Economics and Management, 20(5), 920-938. doi:https://doi.org/10.3846/jbem.2019.10190
56. Sugiartawan, P., Pulungan, R., & Sari, A. K. (2017). Prediction by a Hybrid of Wavelet Transform and Long-Short-Term-Memory Neural Network. International Journal of Advanced Computer Science and Applications, 8(2), 326-332. https://www.researchgate.net/profile/Reza_Pulungan/publication/314166897_Prediction_by_a_Hybrid_of_Wavelet_Transform_and_Long-Short-Term-Memory_Neural_Network/links/5950a68f45851543383a3d3b/Prediction-by-a-Hybrid-of-Wavelet-Transform-and-Long-Short-Term-M.
57. Suresh, K., & Priya, K. (2011). Forecasting sugarcaneyield of Tamilnadu using ARIMA models. Sugar Tech., 23-26.
58. Tsaih, R., Hsu, Y., & Lai, C. (1998). Forecasting S&P 500 Stock Index Futures with A Hybrid Al System. Decision Support Systems(23), 161-174.
59. Tseng, F., Yu, H., & Tzeng, G. (2002). Combining Neural Network Model with Seasonal Time Series ARIMA Model. Tecnological Forecasting&Social Change(69), 71-87.
60. Vasques, X., Vanel, L., Villette, G., & Cif, L. (2016). Morphological Neuron Classification Using Machine Learning. Frontiers in Neuroanatomy, 10, 102-114. doi:10.3389/fnana.2016.00102
61. Wedding, D., & Cios, K. (1996). Time Series Forecasting by Combining RBF Networks, Certainy Factors and The Box-Jenkins Model. Neurocomputing, 10, 149-168.
62. Wu, L., & Brynjolfsson, E. (2015). The Future of Prediction: How Google Searches Foreshadow Housing Prices and Sales. S. M. Avi Goldfarb içinde, Economic Analysis of the Digital Economy (s. 89-118). Chicago, USA: University of Chicago Press. doi:http://dx.doi.org/10.2139/ssrn.2022293
63. Yılmazel, Ö., Afşar, A., & Yılmazel, S. (2018). Konut Fiyat Tahmininde Yapay Sinir Ağları Yönteminin Kullanılması. UİİİD-IJEAS, 285-300.
64. Yu, L., Jiao, C., Xin, H., Wang, Y., & Wang, K. (2018). Prediction on Housing Price Based on Deep Learning. World Academy of Science, Engineering and Technology, 12(2). doi:doi.org/10.5281/zenodo.1315879
65. Yu, L., Wang, S., & Lai, K. (2005). A Novel Nonlinear Ensemble Forecasting Model Incorporating GLAR and ANN for Foreign Exchange Rates. Computers&Operations Research(32), 2523-2541.
66. Zhang, G. (2003). Time Series Forecasting Using a Hybrid ARIMA and Neural Netwok Model. Neurocomputing(50), 159-175. doi:https://doi.org/10.1016/S0925-2312(01)00702-0
67. Zhou, Z., & Hu, C. (2008). An Effective Hybrid Approach Based on Grey and Arma for Forecasting Gyro Drift. Chaos, Solitons and Fractals(35), 525-529.