Prediksi Harga GPU Menggunakan ARIMA Model
DOI:
https://doi.org/10.55382/jurnalpustakaai.v5i3.1420Keywords:
GPU, ARIMA, Peramalan, Deret Waktu, AICAbstract
Graphics Processing Unit (GPU) semakin berperan penting dalam pengembangan kecerdasan buatan, industri game, dan kebutuhan profesional. Namun, harga GPU mengalami volatilitas tinggi dalam beberapa tahun terakhir akibat gangguan rantai pasokan global, aktivitas penambangan kripto, serta meningkatnya permintaan dari pusat data. Peramalan harga GPU yang akurat dapat membantu konsumen dan pelaku bisnis menentukan waktu pembelian yang optimal serta membantu produsen dalam perencanaan produksi dan strategi penetapan harga. Penelitian ini bertujuan untuk meramalkan harga GPU menggunakan pendekatan deret waktu statistik Autoregressive Integrated Moving Average (ARIMA). Dataset yang digunakan berupa harga historis GPU bekas NVIDIA RTX 4090 pada periode Oktober 2022 hingga Juli 2024. Data dibagi secara berurutan menjadi 80% untuk pelatihan dan 20% untuk pengujian. Parameter model diidentifikasi menggunakan uji Augmented Dickey–Fuller (ADF) dan kriteria Akaike Information Criterion (AIC), dengan konfigurasi optimal ARIMA (3,1,3). Hasil evaluasi menunjukkan MAPE = 8,53%, RMSE = 159,75 USD, dan R² = –1,96, yang menunjukkan bahwa ARIMA mampu menangkap tren umum, namun kurang efektif dalam menghadapi pergerakan harga GPU yang sangat fluktuatif. Temuan ini menunjukkan bahwa ARIMA dapat digunakan sebagai model dasar yang andal untuk peramalan jangka pendek, dan dapat ditingkatkan lebih lanjut melalui pendekatan hibrida atau metode berbasis machine learning.
Downloads
References
S. Chauhan, “The Growing Energy Demand of Data Centers: Impacts of AI and Cloud Computing,” International Journal For Multidisciplinary Research, vol. 6, no. 4, Aug. 2024, doi: 10.36948/ijfmr.2024.v06i04.26591.
Emma Rogers, “2025 GPU Prices: 45-55% Above MSRP Amid AI and Crypto Demand,” https://www.webpronews.com/2025-gpu-prices-45-55-above-msrp-amid-ai-and-crypto-demand/, Sep. 01, 2025.
F. Petropoulos et al., “Forecasting: theory and practice,” Jul. 01, 2022, Elsevier B.V. doi: 10.1016/j.ijforecast.2021.11.001.
A. Falkenberg and B. Buchwitz, “Predicting Consumer Goods Prices-The Short-, Medium-And Long-Term Perspective,” IADIS International Journal on Computer Science and Information Systems, vol. 15, no. 2, pp. 58–75.
M. N. Pangestu, M. Jajuli, and U. Enri, “Prediksi Harga Kartu Grafis NVIDIA Berdasarkan Pengaruh Harga Cryptocurrency Menggunakan Support Vector Regression,” Jurnal Ilmiah Wahana Pendidikan, vol. 8, no. 17, pp. 280–287, 2022, doi: 10.5281/zenodo.7076540.
B. D. K. Reddy, J. S. Naik, S. V. Kumar, S. Kumar, G. Haritha, and M. R. Reddy, “A Methodological Review on Time Series Forecasting by using ARIMA,” 2025, pp. 709–719. doi: 10.2991/978-94-6463-662-8_55.
N. Ratna Sari and A. Alfin, “Optimalisasi Rencana Produksi untuk Mengurangi Overstock dan Stockout di Divisi Production Planning and Inventory Control (PPIC) Menggunakan Random Forest,” Jurnal Pustaka AI, vol. 5, pp. 149–165, Aug. 2025, doi: 10.55382/jurnalpustakaai.v5i2.1040.
K. Yeung, “ARIMA Model Application in Predicting NVIDIA’s Stock Price,” Advances in Economics, Management and Political Sciences, vol. 128, pp. 220–230, Dec. 2024, doi: 10.54254/2754-1169/128/2024.18620.
HowMuch.one, “Average Nvidia GeForce RTX 4090 24GB price history,” HowMuch.one. Accessed: Oct. 23, 2025. [Online]. Available: https://howmuch.one/product/average-nvidia-geforce-rtx-4090-24gb/price-history.
J. Wang, J. Hsu, and Z. Qin, “Exploring Nvidia’s Evolution, Innovations, And Future Stock Trends,” Journal of Management and Engineering Integration, vol. 17, no. 1, pp. 21–46, Mar. 2024, doi: 10.62704/10057/28082.
N. Falah Azhari, T. Pasaribu, R. Halomoan Napitupulu, J. Sistem Informasi, F. Sains dan Teknologi, and U. Prima Indonesia, “Pemetaan Tren dan Pola Topik Kekerasan Remaja dengan ARM-TSA pada Pemberitaan Online,” Jurnal Pustaka AI, vol. 5, no. 2, pp. 235–247, 2025, doi: 10.55382/jurnalpustakaai.v5i2.1166.
R. Adhikari and R. K. Agrawal, “An Introductory Study on Time Series Modeling and Forecasting.”
R. J. . Hyndman and George. Athanasopoulos, Forecasting?: principles and practice, Print. Melbourne: OTexts, 2014.
V. Romanuke, “Arima Model Optimal Selection for Time Series Forecasting,” Maritime Technical Journal, vol. 224, no. 1, pp. 28–40, 2022, doi: 10.2478/sjpna-2022-0003.
H. Ben Ameur, S. Boubaker, Z. Ftiti, W. Louhichi, and K. Tissaoui, “Forecasting commodity prices: empirical evidence using deep learning tools,” Ann Oper Res, vol. 339, no. 1–2, pp. 349–367, Aug. 2024, doi: 10.1007/s10479-022-05076-6.
R. Ospina, J. A. M. Gondim, V. Leiva, and C. Castro, “An Overview of Forecast Analysis with ARIMA Models during the COVID-19 Pandemic: Methodology and Case Study in Brazil,” Mathematics, vol. 11, no. 14, 2023, doi: 10.3390/math11143069.
S. Makridakis, E. Spiliotis, and V. Assimakopoulos, “The M4 Competition: 100,000 time series and 61 forecasting methods,” Int J Forecast, vol. 36, no. 1, pp. 54–74, 2020, doi: https://doi.org/10.1016/j.ijforecast.2019.04.014.
D. Chicco, M. J. Warrens, and G. Jurman, “The coefficient of determination R-squared is more informative than SMAPE, MAE, MAPE, MSE and RMSE in regression analysis evaluation,” PeerJ Comput Sci, vol. 7, pp. 1–24, 2021, doi: 10.7717/PEERJ-CS.623.
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2025 Dimas Sakha Fauzi, Fitria Ekarini
This work is licensed under a Creative Commons Attribution 4.0 International License.
