Comparative Analysis of Quantile Autoregression and Prophet Models for IDX30 Index Forecasting
DOI:
https://doi.org/10.36456/jstat.vol19.no1.a11507Keywords:
IDX30, Stock, Forecasting, Quantile Autoregression (QAR), ProphetAbstract
Forecasting stock market indices remains challenging due to volatility, structural breaks, and nonlinear dynamics in financial time series. Although Quantile Autoregression (QAR) and Prophet have been widely applied in forecasting studies, comparative evidence on their performance for the IDX30 index in the Indonesian stock market remains limited. This study aims to compare the forecasting accuracy of QAR and Prophet models in predicting the IDX30 index. Monthly closing price data from May 2012 to March 2026 were divided into an in-sample period (May 2012–March 2020) and an out-of-sample testing period (April 2020–March 2026). A one-step-ahead rolling forecasting approach was employed to evaluate model performance. Forecast accuracy was comprehensively assessed using Symmetric Mean Absolute Percentage Error (sMAPE), Mean Absolute Error (MAE), and Root Mean Squared Error (RMSE). The results demonstrate that the QAR model significantly outperforms Prophet across all evaluation metrics. The QAR model achieved an sMAPE of 3,47%, an RMSE of 21,62, and an MAE of 16,01, which are substantially lower than Prophet’s sMAPE of 8,22%, RMSE of 53,27, and MAE of 38,71. The superior performance of QAR indicates its strength in capturing short-term dependencies and adapting to volatility and structural changes. Practically, QAR offers critical implications for investors and financial analysts; unlike Prophet, which is rigid toward long-term trends, QAR adjusts estimates during extreme market conditions (upper/lower quantiles), leading to more precise and adaptive investment and risk management decisions. These findings confirm that QAR provides a more reliable forecasting framework for the IDX30 index.
References
[1] R. Anwar and L. Rassiyanti, “Analisis Komparasi Model Peramalan Prophet dan ARIMA dalam Memprediksi Harga Saham Penutupan PT ANTM,” vol. 5, no. 1, pp. 57–74, 2025.
[2] R. Koenker and G. Bassett, “Regression Quantiles,” Econometrica, vol. 46, no. 1, pp. 33–50, 1978.
[3] R. Koenker and Z. Xiao, “Quantile Autoregression,” Journal of the American Statistical Association, vol. 101, no. 475, pp. 980–990, 2006.
[4] G. E. P. Box and G. M. Jenkins, Time Series Analysis: Forecasting and Control. San Francisco, CA: Holden-Day, 1976.
[5] S. Makridakis, “The M3-Competition: Results, Conclusions and Implications,” International Journal of Forecasting, vol. 16, no. 4, pp. 451–476, 2000.
[6] J. S. Armstrong, Long-Range Forecasting: From Crystal Ball to Computer. New York, NY: John Wiley & Sons, 1985.
[7] H. Hu, C. Zhao, J. Li, and Y. Huang, “Stock Prediction Model Based on Mixed Fractional Brownian Motion and Improved Fractional-Order Particle Swarm Optimization Algorithm,” Fractal and Fractional, vol. 6, no. 10, 2022.
[8] S. Chavleishvili and S. Manganelli, “Forecasting and Stress Testing with Quantile Vector Autoregression,” Journal of Applied Econometrics, vol. 39, no. 1, pp. 66–85, 2024.
[9] X. Zhang, H. Wang, and Y. Li, “Functional Quantile Autoregression,” Journal of Econometrics, vol. 244, no. 2, 2024.
[10] I. Ghosh and D. Pamucar, “Can Financial Stress Be Anticipated and Explained? Uncovering the Hidden Pattern Using EEMD-LSTM, EEMD-Prophet, and XAI Methodologies,” Complex & Intelligent Systems, vol. 9, pp. 4169–4193, 2023.
[11] A. Sunki, C. SatyaKumar, G. Surya Narayana, and V. Koppera, “Time Series Forecasting of Stock Market Using ARIMA, LSTM and FB Prophet,” MATEC Web of Conferences, vol. 392, 2024.
[12] K. Alim, B. Matsaany, and A. Rahmawati, “Diversification of Jakarta Islamic Index (JII) Stock Optimal Portfolio for the Period 2018–2023,” J Statistika, vol. 16, no. 2, pp. 585–593, 2023.
[13] F. Fitriyani, S. F. A., M. I. R., and T. A. T., “Peramalan Indeks Harga Saham PT Verena Multi Finance Tbk dengan Metode Pemodelan ARIMA dan ARCH-GARCH,” J Statistika, vol. 14, no. 1, pp. 11–23, 2021.
[14] J. Taylor, “Forecasting Value-at-Risk and Expected Shortfall Using Quantile Regression,” International Journal of Forecasting, vol. 40, no. 1, pp. 125–139, 2024.
[15] Y. Li, H. Wang, and X. Zhang, “Quantile-Based Forecasting of Financial Market Risk Under Structural Breaks,” Journal of Forecasting, vol. 43, no. 2, pp. 211–228, 2025.
[16] M. Ahmed and S. Khan, “Comparative Evaluation of Prophet and Machine Learning Models for Financial Time Series Forecasting,” Expert Systems with Applications, vol. 245, 2024.
[17] P. Verma and R. Sharma, “Stock Market Forecasting Using Prophet, LSTM, and Hybrid Approaches: A Comparative Study,” Applied Soft Computing, vol. 152, 2025.
