Journals / Sosyal Bilimler Araştırma Dergisi / 2021 / Cilt: 10 - Sayı: 3

Forecasting of Gold Prices with ARIMA and Random Forest Regression and Evaluating of Their Prediction Performance

ARIMA ve Rassal Orman Yöntemi ile Altın Fiyatlarının Öngörülmesi ve Tahmin Performanslarının Karşılaştırılması

Pages
747–758
Publication Date
2021-09-30
DOI
—
🏛️Bibliographic metadata and full text access for this article are provided via DergiPark.
DergiPark Record ↗
✨

AI Summary & Key Takeaways

AI Insight
⚡ TL;DR(Too Long; Didn't Read)

Therefore, in the direction of aims and outcomes of the study have added an alternative view to gold prices forecasting studies.

📌 3 Key Takeaways

🎯
Objective & Hypothesis:In this study, it was focused on gold price forecasting with time series analysis and machine learning methods aimed to compare forecasting performance.
📊
Key Findings & Data:It was selected for forecasting gold prices that ARIMA (Autoregressive Integrated Moving Average) method from time series analysis and random forest regression type of random forest introduced by Breiman from machine learning method.
💡
Conclusion & Impact:Therefore, in the direction of aims and outcomes of the study have added an alternative view to gold prices forecasting studies.
🔬Design:Modeling & Simulation
📍Scope:Comprehensive Sample
⚙️Variables:Altın Fiyatları · ARIMA · Makine Öğrenmesi

Abstract

In this study, it was focused on gold price forecasting with time series analysis and machine learning methods aimed to compare forecasting performance. It was selected for forecasting gold prices that ARIMA (Autoregressive Integrated Moving Average) method from time series analysis and random forest regression type of random forest introduced by Breiman from machine learning method. In literature, many studies related to gold prices studies in terms of econometrics are onto cointegrating relationship between gold prices and other variables which are affected its. On the other hand, in the gold price forecasting studies with machine learning methods, too little work used Random Forest Regression (RF). In this sense, it is expected to this study which are both using this method and comparing forecasting performance with ARIMA contribute the literature. In analysis, it was separated into two groups are training and test set for each method. The best model which forecast the test set was determined by training set. Root Mean Square Error (RMSE) and Mean Absolute Error (MAE) which are criteria for evaluating forecasting performance was used to determine which method is the best. Result of the analysis indicate that random forest regression is better performance than ARIMA not merely training set but also test set. Therefore, in the direction of aims and outcomes of the study have added an alternative view to gold prices forecasting studies.

Özet

Altın, her dönem insan hayatında önemli bir yer edinmiş değerli bir madendir. Geçmişten günümüze hem değer saklama aracı hem de aksesuar olarak rağbet görmüş bu yatırım aracı günümüzde de değerini korumaya devam etmektedir. Bu çalışmada altın fiyatlarının tahmini üzerine odaklanılmıştır. Tahmin yöntemi olarak, zaman serisi yöntemlerinden ARIMA (Bütünleşik Otoregresif Hareketli Ortalama modeli) ve son dönemde popülerleşen makine öğrenmesi yöntemlerinden Rassal Orman Regresyonu kullanılmıştır. Literatür incelendiğinde altın fiyatlarını rassal orman regresyonu yöntemi ile tahmin eden çok az sayıda çalışma ile karşılaşılmıştır. Bu açıdan, çalışmanın hem rassal orman regresyonu yöntemiyle altın fiyatlarını tahmin etmesi hem de ARIMA tahmini ile tahmin performansını karşılaştırması açısından literatüre katkı sağlayacağı beklenmektedir. Her iki yöntemde de model yapısına bağlı olarak altın fiyatları serisinin geçmiş değerleri kullanılmış ve böylelikle tahmin performansı karşılaştırılmasında tutarlılık sağlanması amaçlanmıştır. Her iki yöntemde de seri, eğitim ve test kümesi olmak üzere iki gruba ayrılmıştır. Eğitim kümesine ilişkin veri kullanılarak uygun kabul edilen model yapısı ile test kümesi tahmin edilmiştir. Literatürde, tahmin performanslarını karşılaştırmak amacıyla yoğun olarak kullanılan RMSE (Root Mean Square Error – Ortalama Hata Karesinin Karekökü) ve MAE (Mean Absolute Error- Ortalama Mutlak Hata) kriterleri kullanıldığından, analiz sonucunda modellerin tahmin performansları sözü edilen kriterlere göre değerlendirilmiş ve her iki kriterde de rassal orman regresyonu yönteminin tahmin açısından daha iyi olduğu sonucuna varılmıştır.

🔖

Cite This Article

DEMİREZEN, S. (2021). Forecasting of Gold Prices with ARIMA and Random Forest Regression and Evaluating of Their Prediction Performance. Sosyal Bilimler Araştırma Dergisi, Cilt: 10 - Sayı: 3, 747–758. https://ajindex.com/articles/801520

📄 Full Text PDF Viewer

📄

PDF görüntüleyici hazırlanıyor...

References (13)

  1. Aye, G., Gupta, R., Hammoudeh, S., & Kim, W. J. (2015). Forecasting the price of gold using dynamic model averaging. International Review of Financial Analysis, 41, 257-266.
  2. Cheung, Y. W., & Lai, K. S. (1997). Bandwidth selection, prewhitening, and the power of the Phillips-Perron test. Econometric Theory, 13(5), 679-691.
  3. Ciaburro, G. (2018). Regression Analysis with R: Design and develop statistical nodes to identify unique relationships within data at scale. Birmingham: Packt Publishing. Cutler, A., Cutler, D. R., & Stevens, J. R. (2012). Random forests. In Ensemble machine learning (pp. 157-175).
  4. Springer, Boston, MA. Dangeti, P. (2017). Statistics for Machine Learning. Birmingham: Packt Publishing. Dickey, D. A., & Fuller, W. A. (1981). Likelihood ratio statistics for autoregressive time series with a unit root. Econometrica: journal of the Econometric Society, 1057-1072.
  5. Dubey, A. D. (2016, January). Gold price prediction using support vector regression and ANFIS models. In 2016 International Conference on Computer Communication and Informatics (ICCCI) (pp. 1-6).
  6. IEEE. Guha, B., & Bandyopadhyay, G. (2016). Gold price forecasting using ARIMA model. Journal of Advanced Management Science, 4(2). Gujarati D. N. & Porter D., Basic Econometrics, ‎ McGraw-Hill Education; 5th edition (October 8, 2008)
  7. Kristjanpoller, W., & Minutolo, M. C. (2015). Gold price volatility: A forecasting approach using the Artificial Neural Network–GARCH model. Expert systems with applications, 42(20), 7245-7251.
  8. Lee, J., & Strazicich, M. C. (2001). Break point estimation and spurious rejections with endogenous unit root tests. Oxford Bulletin of Economics and statistics, 63(5), 535-558.
  9. Lee, J., & Strazicich, M.C. (2004). Minimum LM unit root test with one structural break. Economics Bulletin, 33, 2483-2492.
  10. Leybourne, S. J., & Newbold, P. (1999). The behaviour of Dickey–Fuller and Phillips–Perron testsunder the alternative hypothesis. The Econometrics Journal, 2(1), 92-106.
  11. Makala, D., & Li, Z. (2021, February). Prediction of gold price with ARIMA and SVM. In Journal of Physics: Conference Series (Vol. 1767, No. 1, p. 012022).
  12. IOP Publishing. Manjula, K. A., & Karthikeyan, P. (2019, April). Gold Price Prediction using Ensemble based Machine Learning Techniques. In 2019 3rd International Conference on Trends in Electronics and Informatics (ICOEI) (pp. 1360-1364).
  13. IEEE. Mirmirani, S., & Li, H. C. (2004). Gold price, neural networks and genetic algorithm. Computational Economics, 2