Peramalan Menggunakan Model Hybrid ARIMAX-NN untuk Total Transaksi Pembayaran Nontunai

Nuning Kusumaningrum; Ika  Purnamasari; Meiliyani  Siringoringo

doi:10.35580/variansiunm57

Nuning Kusumaningrum Universitas Mulawarman
Ika Purnamasari Jurusan Matematika, Fakultas Matematika dan Ilmu Pengetahuan Alam, Mulawarman University,
Meiliyani Siringoringo Jurusan Matematika, Fakultas Matematika dan Ilmu Pengetahuan Alam, Mulawarman University,

DOI: https://doi.org/10.35580/variansiunm57

Keywords: ARIMAX, Forecasting, Hybrid, Neural Network, Non-cash transactions

Abstract

Non-cash payment transactions in Indonesia continue to experience an increase marked by the high consumptive behavior of the people. This consumptive behavior is based on the many attractive offers, especially on year-end holidays which are the effect of calendar variations. ARIMAX is a time series method that is able to detect the effects of calendar variations. Meanwhile, to increase the level of forecasting accuracy, it can be combined with other methods such as Neural Networks (NN). This study aims to predict the total non-cash payment transactions in Indonesia in the period January to December 2022 using the ARIMAX-NN hybrid model. Based on the forecasting results, four highly accurate models were obtained, namely the hybrid model ARIMAX(0,1,2)-NN 1 neuron, ARIMAX(0,1,2)-NN 2 neurons, ARIMAX(1,1,0)-NN 1 neurons, and ARIMAX(1,1,0)-NN 2 neurons with MAPE values for each model below 5%. Based on the four models formed, the results of forecasting in the period January to December 2022 as a whole the data tends to fluctuate and has an upward trend pattern, especially in December, which is the month when year-end holidays occur.

References

Alam, W., Mrinmoy, R. A. Y., Kumar, R. R., Sinha, K., Rathod, S., & Singh, K. N. (2018). Improved ARIMAX Model Based on ANN and SVM Approaches for Forecasting Rice Yield Using Weather Variables. Indian Journal of Agricultural Sciences, 88(12), 1909–1913.

Amalia, F. F., Suhartono, Rahayu, S. P., & Suhermi, N. (2018). Quantile Regression Neural Network for Forecasting Inflow and Outflow in Yogyakarta. Journal of Physics: Conference Series, 1028, 1–10.

Bennett, C., Stewart, R. A., & Lu, J. (2014). Autoregressive with Exogenous Variables and Neural Network Short-term Load Forecast Models for Residential Low Voltage Distribution Networks. Energies, 7(5), 2938–2960.

Berlinditya, O. R. E. . B., & Noeryanti. (2019). Pemodelan Time Series Dalam Peramalan Jumlah Pengunjung Objek Wisata Di Kabupaten Gunung Kidul Menggunakan Metode ARIMAX Efek Variasi Kalender. Jurnal Statistika Industri dan Komputasi, 4(1), 81–88.

Farasyi, F. Al, & Iswati, H. (2021). Pengaruh Media Sosial, E-Lifestyle dan Budaya Digital Terhadap Perilaku Konsumtif. Syntax Idea, 3(11), 2355–2371.

Febriaty, H. (2019). Pengaruh Sistem Pembayaran Non Tunai Dalam Era Digital Terhadap Tingkat Pertumbuhan Ekonomi Indonesia. Prosiding FRIMA (Festival Riset Ilmiah Manajemen dan Akuntansi), 6681(2), 307–313.

Hudiyanti, C. V., Bachtiar, F. A., & Setiawan, B. D. (2019). Perbandingan Double Moving Average dan Double Exponential Smoothing untuk Peramalan Jumlah Kedatangan Wisatawan Mancanegara di Bandara Ngurah Rai. Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer, 3(3), 2667–2672.

Intan, S. N., Zukhronah, E., & Wibowo, S. (2019). Peramalan Banyaknya Pengunjung Pantai Glagah Menggunakan Metode Autoregressive Integrated Moving Average Exogenous (ARIMAX) dengan Efek Variasi Kalender. Indonesian Journal of Applied Statistics, 1(2), 70–78.

Makridakis, S., Wheelwright, S. C., & McGee, V. E. (1999). Metode dan Aplikasi Peramalan, Jilid 1. Jakarta: Erlangga.

Lee, M. H., & Hamzah, Suhartono, N. A. (2010). Calendar variation model based on ARIMAX for forecasting sales data with Ramadhan effect Calendar variation model based on ARIMAX for forecasting sales data with Ramadhan effect. Proceeding of the Regional Conference on Statistical Sciences 2010 (RCSS’10), 349–361.

Meliana, C., Wasono, R., Al Haris, M., Alfiyani, Z. H., & Sari, E. Y. K. (2020). Peramalan Indeks Harga Saham Gabungan (IHSG) Mengguanakn ARIMAX dengan Variabel Eksogen Covid-19. Prosiding Seminar Edusaintech, 258–267.

Nursari, A., Suparta, I. W., & Moelgini, Y. (2019). Pengaruh Pembayaran Non Tunai Terhadap Jumlah Uang yang Diminta Masyarakat (M1) dan Perekonomian. JEP, 8(3), 285–306.

Prastyaningtyas, E. W. (2019). Dampak Ekonomi Digital Bagi Perekonomian Indonesia. Seminar Nasional Manajemen Ekonomi dan Akuntansi (SENMEA), IV, 103–108.

Prastyo, D. D., Suhartono, Puka, A. O. B., & Lee, M. H. (2018). Comparison Between Hybrid Quantile Regression Neural Network and Autoregressive Integrated Moving Average with Exogenous Variable for Forecasting of Currency Inflow and Outflow in Bank Indonesia. Jurnal Teknologi, 80(6), 61–68.

Putera, M. L. S. (2020). Peramalan Transaksi Pembayaran Non-Tunai Menggunakan ARIMAX-ANN dengan Konfigurasi Kalender. Jurnal Ilmu Matematika dan Terapan, 14(1), 135–146.

Rosadi, D. (2012). Ekonometrika & Analisis Runtun Waktu Terapan dengan Eviews. Yogyakarta: Penerbit ANDI.

Warsito, B. (2009). Kapita Selekta Statistika Neural Network. Semarang: BP UNDIP.

Wei, W. W. S. (2006). Time Series Analysis: Univariate and Multivariate Methods Second Edition. Boston: Pearson Education, Inc.

Zhang, P. G. (2003). Time series forecasting using a hybrid ARIMA and neural network model. Neurocomputing, 50, 159–175.

Zhang, P. G. (2004). Neural Network in Busines Forecasting. Hershy: Idea Group Publishing.