Comparative Analysis of Local Polynomial Regression and ARIMA in Predicting Indonesian Benchmark Coal Price

Arinda Mahadesyawardani; Utsna Rosalin Maulidya; Barnabas Anthony Philbert Marbun; Fachriza Yosa Pratama; Nur Chamidah

doi:10.21831/pythagoras.v19i1.74889

Comparative Analysis of Local Polynomial Regression and ARIMA in Predicting Indonesian Benchmark Coal Price

Arinda Mahadesyawardani, Department of Statistics, Universitas Airlangga, Surabaya, Indonesia, Indonesia
Utsna Rosalin Maulidya, Department of Statistics, Universitas Airlangga, Surabaya, Indonesia, Indonesia
Barnabas Anthony Philbert Marbun, Department of Statistics, Universitas Airlangga, Surabaya, Indonesia, Indonesia
Fachriza Yosa Pratama, Department of Statistics, Universitas Airlangga, Surabaya, Indonesia, Indonesia
Nur Chamidah, Department of Statistics, Universitas Airlangga, Surabaya, Indonesia, Indonesia

10.21831/pythagoras.v19i1.74889

Abstract

As one of the world's biggest coal producers, it is essential for Indonesia to follow the trend of benchmark coal price fluctuations for any future possibilities. This study compared two methods of forecasting benchmark coal prices to evaluate the accuracy of the predictions used a nonparametric regression based on the local polynomial estimator and a parametric ARIMA method. Local polynomial analysis obtained a MAPE of 2.929278% using a CV method based on optimal bandwidth of 5.06 at order 2 with a cosine kernel, which means highly accurate forecasting accuracy. As for the ARIMA analysis, the data does not meet the assumption of normality, but forecasting is still continued with the best model ARIMA (1,2,1) model so that the MAPE is 12.6327%, which means good forecasting accuracy. Therefore in this study, the use of nonparametric regression methods using local polynomial estimators on data with non-normal distribution are more suitable to obtain accurate prediction results.

Keywords

nonparametric regression; local polynomial; cross validation; ARIMA; benchmark coal price

References

Ardi, T., Santoso, R., & Prahutama, A. (2017). Implementasi Subset Autoregressive Menggunakan Paket Fitar. Jurnal Gaussian, 6(4), 510–519. https://doi.org/10.14710/j.gauss.6.4.510-519

Ardianti, C. W., Santoso, R., & Sudarno, S. (2020). Analisis Arima Dan Wavelet Untuk Peramalan Harga Cabai Merah Besar Di Jawa Tengah. Jurnal Gaussian, 9(3), 247–262. https://doi.org/10.14710/j.gauss.v9i3.28906

Chamidah, N., & Lestari, B. (2022). Analisis Regresi Nonparametrik dengan Perangkat Lunak R. Surabaya: Airlangga University Press.

Chang, P. C., Y.W, W., & C.H, L. (2007). The Development of a Weighted Evolving Fuzzy Neural Network for PCB Sales Forecasting. Journal Expert System with Applications, 32(1), 86–96. https://doi.org/10.1016/j.eswa.2005.11.021

Gunarto, M., & Wulansari, R. (2020). Analisis pergerakan harga saham berdasarkan sarga scuan dan volume penjualan : Ssudi pada PT Bukit Asam Tbk. Jurnal Manajemen Dan Bisnis Sriwijaya, 18(4), 255-272. https://doi.org/10.29259/jmbs.v18i4.13021

Hendrawan, B. (2013). Penerapan Model ARIMA Dalam Memprediksi IHSG. Politeknik Negeri Batam.

Hidayanti, R. P., Mustikasari, & Hariani. (2022). Prediksi Harga Batu Bara Menggunakan Regresi. Journal of Embedded System Security and Intelligent System, 3(1), 1-10. https://doi.org/10.26858/jessi.v3i1.33142

IEA. (2023). Coal 2023. International Energy Agency. https://www.iea.org/reports/coal-2023/executive-summary

Kementerian ESDM. (2024). Harga Acuan. Direktorat Jenderal Mineral dan Batubara. https://www.minerba.esdm.go.id/harga_acuan

Kementerian ESDM. (2022). Laporan Kinerja 2022. Direktorat Jenderal Mineral dan Batubara. https://www.minerba.esdm.go.id/upload/file_menu/20230329164006.pdf

Kementerian ESDM. (in press). Produksi Batubara Domestik Tembus Target, Ketahanan Energi Nasional Terjaga. Kementerian ESDM RI. https://www.esdm.go.id/id/media-center/arsip-berita/produksi-batubara-domestik-tembus-target-ketahanan-energi-nasional-terjaga

Kurniasih, D., Mariani, S., & Sugiman. (2013). Efisiensi Relatif Estimator Fungsi Kernel Gaussian Terhadap Estimator Polinomial Dalam Peramalan Usd Terhadap Jpy. UNNES Journal of Mathematics, 2, 79–84.

Makridakis, S., WheelWright, S., & McGree, V, E. (1999). Metode dan Aplikasi Peramalan (2nd ed.). Jakarta: Erlangga.

Pitaloka, R. A., Sugito, S., & Rahmawati, R. (2019). Perbandingan Metode Arima Box-Jenkins Dengan ARIMA Ensemble Pada Peramalan Nilai Impor Provinsi Jawa Tengah. Jurnal Gaussian, 8(2), 194–207. https://doi.org/10.14710/j.gauss.v8i2.26648

Prahutama, A., Suparti, D. I., & Utami, T. W. (2018). Pemodelan Bivariate Polinomial Lokal Pada Jumlah Kematian Ibu Dan Bayi Di Jawa Tengah. Prosiding Seminar Nasional Venue Artikulasi-Riset, Inovasi, Resonansi-Teori, Dan Aplikasi Statistika (VARIANSI), 2018(Vol 1), 209–220. http://ojs.unm.ac.id/variansistatistika/article/view/7208

Salim, M. I., Adnan Sauddin, & M. Ichsan Nawawi. (2022). Model Regresi Nonparametrik Deret Fourier Pada Kasus Tingkat Pengangguran Terbuka Di Sulawesi Selatan. Jurnal MSA ( Matematika Dan Statistika Serta Aplikasinya), 10(2), 48–56. https://doi.org/10.24252/msa.v10i2.30993

Suci Pujiani Prahesti, Itasia Dina Sulvianti, & Yenni Angraini. (2023). Peramalan Harga Batu Bara Acuan Menggunakan Metode Autoregressive Integrated Moving Average Dan Fungsi Transfer. Xplore: Journal of Statistics, 12(1), 1–11. https://doi.org/10.29244/xplore.v12i1.1100

Shantika Martha, N. A. N. N. D. (2019). Estimasi Model Regresi Nonparametrik Kernel Menggunakan Estimator Nadaraya-Watson. Bimaster : Buletin Ilmiah Matematika, Statistika Dan Terapannya, 8(4), 633–638. https://doi.org/10.26418/bbimst.v8i4.35870

Suparti, S., & Santoso, R. (2024). Analisis Data Time Series Menggunakan Model Kernel: Pemodelan Data Harga Saham MDKA. Indonesian Journal of Applied Statistics, 6(1), 22. https://doi.org/10.13057/ijas.v6i1.79385

Wei, W. W. S. (2006). Time Series Analysis:Univariate and Multivariate Methods. New York: Pearson.

Welsh, A. H., & Yee, T. Y. (2006). Local Regression for Vector Responses. Journal of Statistical Planning and Inference, 136(9), 3007–3031. https://doi.org/10.1016/j.jspi.2004.01.024

DOI: https://doi.org/10.21831/pythagoras.v19i1.74889