Penerapan Metode ARIMAX dengan Efek Variasi Kalender pada Peramalan Harga Komoditas Cabai Rawit di Provinsi Jawa Barat
Abstract
Abstract. Forecasting is an important aid in planning and decision-making. One of the forecasting methods that is often used is the time series method. Time series are often influenced by a particular event or other variable, so it can cause the data to have a different repeating pattern each period. The ARIMAX Model is thought to be able to capture certain patterns by including certain event information as additional variables, or so-called exogenous variables. The tendency of data to show patterns at certain times based on dates in the calendar is called calendar variation, and the effect of calendar variation can appear on certain events such as Eid al-Fitr. Special treatment is needed for time series data with a calendar variation effect, where the ARIMAX model is well applied to the series data in that case, so that the model formed is ARIMAX with calendar variation. In this thesis, I conducted a study on forecasting the price of cayenne pepper in West Java province using the ARIMAX model with a variation of the calendar in which to include information on Eid al-Fitr events as an additional variable. The best ARIMAX model for forecasting is ARIMAX (0, 1, 1) with accuracy using a MAPE value of 11%, which is based on the criteria of forecasting ability. It can be concluded that the model ARIMAX (0, 1, 1) has good forecasting ability.
Abstrak. Peramalan merupakan bantuan penting dalam perencanaan dan pengambilan keputusan. Salah satu metode peramalan yang sering digunakan adalah metode deret waktu (time series). Pada deret waktu seringkali dipengaruhi oleh suatu peristiwa tertentu atau variabel lain, sehingga dapat menyebabkan data memiliki pola berulang berbeda setiap periodenya. Model ARIMAX diduga mampu menangkap pola tertentu dengan memasukkan informasi peristiwa tertentu sebagai variabel tambahan atau disebut variabel eksogen. Adanya kecenderungan data untuk menampilkan pola pada waktu tertentu berdasarkan penanggalan di dalam kalender disebut variasi kalender, dimana efek variasi kalender dapat muncul pada peristiwa tertentu seperti Idul Fitri. Diperlukan perlakuan khusus untuk data deret waktu dengan efek variasi kalender, dimana model ARIMAX baik diterapkan untuk data deret dengan kasus tersebut, sehingga model yang terbentuk adalah ARIMAX dengan variasi kalender. Dalam skripsi ini melakukan penelitian pada peramalan harga cabai rawit di Provinsi Jawa Barat menggunakan model ARIMAX dengan variasi kalender dimana memasukkan informasi peristiwa Idul Fitri sebagai variabel tambahan. Didapatkan model ARIMAX terbaik untuk melakukan peramalan yaitu ARIMAX(0,1,1) dengan akurasi menggunakan nilai MAPE sebesar 11%, dimana berdasarkan kriteria kemampuan peramalan bahwa dapat disimpulkan model ARIMAX(0,1,1) memiliki kemampuan peramalan yang baik.
References
Alagidede, P. (2008). Month-of-the-year and Pre-Holiday Seasonality In African Stock Markets. Stirling Economics Discussion Paper.
Azizah, N. (2023). Pemodelan Spatial Autoregressive (SAR-X) pada Perkawinan Usia Anak di Indonesia. Jurnal Riset Statistika, 1–10. https://doi.org/10.29313/jrs.v3i1.1643
Chang, P. C., Wang, Y. W., & Liu, C. H. (2007). The Development of a Weighted Evolving Fuzzy Neural Network for PCB Sales Forecasting. Expert Systems with Applications, 32(1), 86–96. https://doi.org/10.1016/j.eswa.2005.11.021
Cryer, J. D., & Chan, K.-S. (2008). Time Series Analysis with Application in R, 2nd Edition. New York: Springer.
Hyndman, R. J., & Athanasopoulos, G. (2018). Forecasting: Principles and Practice. Monash University.
Liu, L.-M. (1980). Note—Analysis of Time Series with Calendar Effects. Management Science, 26(1), 106–112. https://doi.org/10.1287/mnsc.26.1.106
Makridakis, S., Wheelwright, S. C., & Hyndman, R. J. (1997). Forecasting Methods and Applications, 3rd. New York: John Wiley & Sons.
Mulyana. (2004). Buku Ajar Analisis Data Deret Waktu. In Universitas Padjadjaran FMIPA Jurusan. Statistika, Bandung.
Putranto, W., Sayekti, S., Indrayanti, S., & Suhariyanto. (2011). Bunga Rampai Statistik Percabaian. Jakarta: Badan Pusat Statistika.
Setiawan, Suhartono, Ahmad, I. S., & Rahmawati, N. I. (2015). Configuring Calendar Variation Based on Time Series Regression Method for Forecasting of Monthly Currency Inflow and Outflow in Central Java. AIP Conference Proceedings, 1691. https://doi.org/10.1063/1.4937106
Shumway, R. H., & Stoffer, D. S. (2016). Time Series Analysis and Its Applications With R Examples 4th Edition. New York: Spinger.
Siswanti, T. E., & Yanti, T. S. (2020). Pemodelan ARIMAX (Autoregressive Integrated Moving Average with Exogenous Variable). Prosiding Statistika, 6(2), 113–118.
Sudrajat, A. (2019). Disperindag Jabar Kendalikan Harga Kebutuhan Pokok Melalui Pemantauan Pasar. Antara Jabar. https://jabar.antaranews.com/berita/125488/disperindag-jabar-kendalikan-harga-kebutuhan-pokok-melalui-pemantauan-pasar
Susilawati, R., & Sunendiari, S. (2022). Peramalan Jumlah Penumpang Kereta Api Menggunakan Metode Arima dan Grey System Theory. Jurnal Riset Statistika, 1–13. https://doi.org/10.29313/jrs.vi.603
Syam, A. R. P. (2022). Application of the Autoregressive Integrated Moving Average Exogenous (ARIMAX) with Calendar Variation Effect Method for Forecasting Chocolate Data in Indonesia and the United States. Jurnal Matematika, Statistika Dan Komputasi, 18(2), 224–236. https://doi.org/10.20956/j.v18i2.18460
Umar, M. A. (2018). Penerapan Model ARIMAX pada Data Harga Cabai Merah Keriting di Indonesia. Institut Pertanian Bogor.
Wei, W. W. (2006). Time Series Analysis: Univariate and Multivariate Methods, 2nd Edition. New York: Pearson Education, Inc.
Yanti, T. S. (2009). Model Pengganda Uang untuk Menentukan Jumlah Uang Beredar di Indonesia Menggunakan Model ARIMA Komponen. Statistika, 9(1), 25–32.
