Comparison of Time Series Techniques for Predicting Gold and Oil Prices
Article Sidebar
Published:
Aug 31, 2018
Keywords:
Oil Prices; Gold Prices; Time Series Techniques
Main Article Content
Abstract
Forecasting oil and gold prices from a word market prices reference is a challenging
research task. By accurately forecasting the price of gold, it is especially useful for investors.
Time series techniques play a role in predicting future time series data. Therefore, this research
aims to compare performance of time series techniques for predicting gold and crude oil prices.
The data were collected from 2 January 2003 to 30 December 2016. In this research,
six techniques including Linear Regression (LR), Multi-Layer Perceptron (MLP), Radial
Basis Function (RBF), Reduced Error Pruning Tree (REPT), Support Vector Machine
Regression (SVMR) and Sequential Minimal Optimization Regression (SMOR) were used.
Sliding Windows was used to divide data into learning and testing sets. 13 rounds of sliding
windows were used to reduce the variance of experiment results. Moreover, Mean Absolute
Error (MAE) and Root Mean Square Error (RMSE) were used to evaluate the performance
of the model. This study found that the SMOR technique is eff ective in forecasting gold
and oil prices with the lowest MAE values at 14.21±5.35 and 1.65±0.75.
Article Details
How to Cite
[1]
ญึก ว. and ทองคำ จ., “Comparison of Time Series Techniques for Predicting Gold and Oil Prices”, RMUTI Journal, vol. 11, no. 2, pp. 154–167, Aug. 2018.
Section
Research article
References
[1] Khashman, A. and Nwulu, N. I. (2011). Support Vector Machines Versus Back Propagation
Algorithm for Oil Price Prediction. 8th International Symposium on Neural Networks,
ISNN 2011 Guilin, China, May/June 2011 Proceedings, Part |||. Vol. 6677, Liu, D.,
Z.H., Polycarpou M., Alippi C., He H. (eds) Ed., ed: Springer, Berlin, Heidelberg, pp. 530-538
Round11 1.85 ±0.41 4.58 ±1.96 40.23 ±0.19 40.20 ±0.20 23.42 ±0.20 2.58 ±0.72
Round12 3.43 ±1.58 8.76 ±3.84 32.77 ±0.34 32.78 ±0.43 16.18 ±0.44 4.37 ±2.23
Round13 0.72 ±0.15 0.96 ±0.27 31.40 ±0.35 31.40 ±0.35 15.08 ±0.35 2.19 ±1.07
Average 1.88 ±1.12 3.87 ±2.74 40.74 ±4.57 40.71 ±4.53 22.31 ±0.32 1.91 ±0.73
Rounds LR MLP RBF REPT SVMR SMOR
166 การเปรียบเทียบเทคนิคอนุกรมเวลาเพื่อพยากรณ์ราคาทองและราคานํ้ามัน
[2] Dubey, A. D. (2016). Gold Price Prediction Using Support Vector Regression and ANFIS
Models. 2016 International Conference on Computer Communication and Informatics
(ICCCI). pp. 1-6. DOI: 10.1109/ICCCI.2016.7479929
[3] Zainal, N. A. and Mustaff a, Z. (2016). Developing A Gold Price Predictive Analysis Using Grey
Wolf Optimizer. 2016 IEEE Student Conference on Research and Development (SCOReD).
pp. 1-6. DOI: 10.1109/SCORED.2016.7810031
[4] Christina, C. and Umbara, R. F. (2015). Gold Price Prediction Using Type-2 Neuro-Fuzzy Modeling
and ARIMA. 2015 3rd International Conference on Information and Communication
Technology (ICoICT). pp. 272-277. DOI: 10.1109/ICoICT.2015.7231435
[5] Kanokkarn, M. (2014). Study of the Appropriate Forecasting Methods for Consumer Product
Demand of a Public Company. Journal of Business Administration The Association of
Private Education Institutions of Thailand. Vol. 3, No. 1, pp. 12-21
[6] Mombeini, H. and Yazdani-Chamzini, A. (2015). Modeling Gold Price Via Artifi cial Neural
Network. Journal of Economics, Business and Management, Vol. 3, No.7, pp. 699-703. DOI:
10.7763/JOEBM.2015.V3.269
[7] Yang, J. -F., Zhai, Y. -J., Xu, D. -P., and Han, P. (2007). SMO Algorithm Applied in Time Series
Model Building and Forecast. 2007 International Conference on Machine Learning and
Cybernetics. pp. 2395-2400
[8] World Gold Council. (2016). The Market Development Organisation for the Gold Industry.
Access (30 January 2017). Available (https://www.gold.org)
[9] Organization of the Petroleum Exporting Countries. (2016). Monthly Oil Market Report.
Access (30 April 2017). Available (https://www.opec.org/opec_web/en/)
[10] Ongsritrakul, P. and Soonthornphisaj, N. (2003). Apply Decision Tree and Support Vector
Regression to Predict the Gold Price. Proceedings of the International Joint Conference
on Neural Networks, 2003. Vol. 4, pp. 2488-2492. DOI: 10.1109/IJCNN.2003.1223955
[11] KangaraniFarahani, M. and Mehralian, S., (2013). Comparison Between Artifi cial Neural
Network and Neuro-Fuzzy for Gold Price Prediction. 2013 13th Iranian Conference on
Fuzzy Systems (IFSC). pp. 1-5. DOI: 10.1109/IFSC.2013.6675635
[12] Xie, W., Yu, L., ShanyingXu, and Wang, S. (2006). A New Method for Crude Oil Price Forecasting
Based on Support Vector Machines. International Conference on Computational Science.
Computational Science- ICCS 2006. pp. 444-451. DOI: 10.1007/11758549_63
[13] Nwulu, N. I. (2017). A Decision Trees Approach to Oil Price Prediction. 2017 International
Artifi cial Intelligence and Data Processing Symposium (IDAP). pp. 1-5. DOI: 10.1109/IDAP.
2017.8090313
วารสาร มทร.อีสาน ฉบับวิทยาศาสตร์และเทคโนโลยี ปีที่ 11 ฉบับที่ 2 พฤษภาคม - สิงหาคม 2561 167
[14 Peña, M. A., Brenning, A., and Liao, R., (2017). Classifying Fruit-Tree Crops by Landsat-8 time
series. 2017 First IEEE International Symposium of Geoscience and Remote Sensing
(GRSS-CHILE). pp. 1-4. DOI: 10.1109/GRSS-CHILE.2017.7995998
[15] Frank, E. (2014). Fully Supervised Training of Gaussian Radial Basis Function Networks
in WEKA. Department of Computer Science University of Waikato.
[16] Elomaa, T. and Kaariainen, M. (2001). An Analysis of Reduced Error Pruning. Journal of
Artifi cial Intelligence Research. Vol. 15, Issue 1, pp. 163-187
[17] Chang, C. -C. and Lin, C. -J. (2011). LIBSVM : A Library for Support Vector Machines. ACM
Transactions on Intelligent Systems and Technology (TIST). Vol. 2, Issue 3, pp. 1-27. DOI:
10.1145/1961189.1961199
[18] Shevade, S. K., Keerthi, S. S., Bhattacharyya, C., and Murthy, K. R. K. (2000). Improvements
to the SMO Algorithm for SVM Regression. IEEE Transactions on Neural Networks. Vol. 11,
No. 5, pp. 1188-1193. DOI: 10.1109/72.870050
Algorithm for Oil Price Prediction. 8th International Symposium on Neural Networks,
ISNN 2011 Guilin, China, May/June 2011 Proceedings, Part |||. Vol. 6677, Liu, D.,
Z.H., Polycarpou M., Alippi C., He H. (eds) Ed., ed: Springer, Berlin, Heidelberg, pp. 530-538
Round11 1.85 ±0.41 4.58 ±1.96 40.23 ±0.19 40.20 ±0.20 23.42 ±0.20 2.58 ±0.72
Round12 3.43 ±1.58 8.76 ±3.84 32.77 ±0.34 32.78 ±0.43 16.18 ±0.44 4.37 ±2.23
Round13 0.72 ±0.15 0.96 ±0.27 31.40 ±0.35 31.40 ±0.35 15.08 ±0.35 2.19 ±1.07
Average 1.88 ±1.12 3.87 ±2.74 40.74 ±4.57 40.71 ±4.53 22.31 ±0.32 1.91 ±0.73
Rounds LR MLP RBF REPT SVMR SMOR
166 การเปรียบเทียบเทคนิคอนุกรมเวลาเพื่อพยากรณ์ราคาทองและราคานํ้ามัน
[2] Dubey, A. D. (2016). Gold Price Prediction Using Support Vector Regression and ANFIS
Models. 2016 International Conference on Computer Communication and Informatics
(ICCCI). pp. 1-6. DOI: 10.1109/ICCCI.2016.7479929
[3] Zainal, N. A. and Mustaff a, Z. (2016). Developing A Gold Price Predictive Analysis Using Grey
Wolf Optimizer. 2016 IEEE Student Conference on Research and Development (SCOReD).
pp. 1-6. DOI: 10.1109/SCORED.2016.7810031
[4] Christina, C. and Umbara, R. F. (2015). Gold Price Prediction Using Type-2 Neuro-Fuzzy Modeling
and ARIMA. 2015 3rd International Conference on Information and Communication
Technology (ICoICT). pp. 272-277. DOI: 10.1109/ICoICT.2015.7231435
[5] Kanokkarn, M. (2014). Study of the Appropriate Forecasting Methods for Consumer Product
Demand of a Public Company. Journal of Business Administration The Association of
Private Education Institutions of Thailand. Vol. 3, No. 1, pp. 12-21
[6] Mombeini, H. and Yazdani-Chamzini, A. (2015). Modeling Gold Price Via Artifi cial Neural
Network. Journal of Economics, Business and Management, Vol. 3, No.7, pp. 699-703. DOI:
10.7763/JOEBM.2015.V3.269
[7] Yang, J. -F., Zhai, Y. -J., Xu, D. -P., and Han, P. (2007). SMO Algorithm Applied in Time Series
Model Building and Forecast. 2007 International Conference on Machine Learning and
Cybernetics. pp. 2395-2400
[8] World Gold Council. (2016). The Market Development Organisation for the Gold Industry.
Access (30 January 2017). Available (https://www.gold.org)
[9] Organization of the Petroleum Exporting Countries. (2016). Monthly Oil Market Report.
Access (30 April 2017). Available (https://www.opec.org/opec_web/en/)
[10] Ongsritrakul, P. and Soonthornphisaj, N. (2003). Apply Decision Tree and Support Vector
Regression to Predict the Gold Price. Proceedings of the International Joint Conference
on Neural Networks, 2003. Vol. 4, pp. 2488-2492. DOI: 10.1109/IJCNN.2003.1223955
[11] KangaraniFarahani, M. and Mehralian, S., (2013). Comparison Between Artifi cial Neural
Network and Neuro-Fuzzy for Gold Price Prediction. 2013 13th Iranian Conference on
Fuzzy Systems (IFSC). pp. 1-5. DOI: 10.1109/IFSC.2013.6675635
[12] Xie, W., Yu, L., ShanyingXu, and Wang, S. (2006). A New Method for Crude Oil Price Forecasting
Based on Support Vector Machines. International Conference on Computational Science.
Computational Science- ICCS 2006. pp. 444-451. DOI: 10.1007/11758549_63
[13] Nwulu, N. I. (2017). A Decision Trees Approach to Oil Price Prediction. 2017 International
Artifi cial Intelligence and Data Processing Symposium (IDAP). pp. 1-5. DOI: 10.1109/IDAP.
2017.8090313
วารสาร มทร.อีสาน ฉบับวิทยาศาสตร์และเทคโนโลยี ปีที่ 11 ฉบับที่ 2 พฤษภาคม - สิงหาคม 2561 167
[14 Peña, M. A., Brenning, A., and Liao, R., (2017). Classifying Fruit-Tree Crops by Landsat-8 time
series. 2017 First IEEE International Symposium of Geoscience and Remote Sensing
(GRSS-CHILE). pp. 1-4. DOI: 10.1109/GRSS-CHILE.2017.7995998
[15] Frank, E. (2014). Fully Supervised Training of Gaussian Radial Basis Function Networks
in WEKA. Department of Computer Science University of Waikato.
[16] Elomaa, T. and Kaariainen, M. (2001). An Analysis of Reduced Error Pruning. Journal of
Artifi cial Intelligence Research. Vol. 15, Issue 1, pp. 163-187
[17] Chang, C. -C. and Lin, C. -J. (2011). LIBSVM : A Library for Support Vector Machines. ACM
Transactions on Intelligent Systems and Technology (TIST). Vol. 2, Issue 3, pp. 1-27. DOI:
10.1145/1961189.1961199
[18] Shevade, S. K., Keerthi, S. S., Bhattacharyya, C., and Murthy, K. R. K. (2000). Improvements
to the SMO Algorithm for SVM Regression. IEEE Transactions on Neural Networks. Vol. 11,
No. 5, pp. 1188-1193. DOI: 10.1109/72.870050