Research Article
BibTex RIS Cite

XGBOOST VE MARS YÖNTEMLERİYLE ALTIN FİYATLARININ KESTİRİMİ

Year 2020, Issue: 83, 427 - 446, 30.09.2020

Abstract

Altın önemli bir ödeme, yatırım ve birikim aracı olduğundan fiyatının belirlenmesi ülkeler ve yatırımcılar için önemlidir. Bu nedenle bu çalışmada altın fiyatının kestirimi amaçlanmıştır. Bu amaçla altın fiyatı üzerinde etkili olduğu düşünülen gümüş fiyatı, ham petrol WTI vadeli işlemleri fiyatı, ABD Doları endeksi, S&P500 endeksi, ABD federal fonlar bileşik faiz oranı, ABD TÜFE değişkenleri oluşturulan modellerde girdi olarak kullanılmıştır. Kullanılan veriler Ocak 2015 – Haziran 2020 dönemine aittir. Altın fiyatı doğrusal olmayan bir seridir, bunun yanında durağandışıdır. Altın fiyatının bu özellikleri fiyat kestirimlerin elde edilmesini zorlaştırmaktadır. Bu nedenle klasik yöntemlerin yanında makine öğrenmesi yöntemlerinin ve parametrik olmayan yöntemlerin altın fiyatının kestiriminde kullanılması uygun olmaktadır. Bu çalışmada, kestirimlerin elde edilmesinde XGBoost, MARS ve lineer regresyon modelleri kullanılmıştır. Elde edilen sonuçlar modellere ait performans değerlendirme kriterleri kullanılarak karşılaştırılmış, XGBoost ve MARS modelleri için girdi değişkenlerin altın fiyatı üzerindeki etkileri belirlenmiştir. Kullanılan modeller arasında XGBoost modeli %99,6 başarılı kestirim oranı ile en başarılı sonuçların elde edilmesini sağlamıştır. MARS modeli için ise bu oran %97,8’dir. Bu oranlar kullanılan değişkenlerin altın fiyatı üzerinde önemli etkiye sahip olduğunu göstermektedir. Kullanılan değişkenler arasında altın fiyatı üzerinde en önemli etkiye sahip değişken ABD TÜFE değişkenidir. Ayrıca elde edilen bulgular XGBoost ve MARS yöntemlerinin altın fiyatı ve benzer seriler için kestirimlerin elde edilmesinde tercih edilebilecek yöntemler olduğunu göstermektedir.

References

  • Akin, M., Eyduran, S. P., Eyduran, E. ve Reed, B. M. (2020). Analysis of macro nutrient related growth responses using multivariate adaptive regression splines. Plant Cell, Tissue and Organ Culture (PCTOC), 140(3), 661-670.
  • Alameer, Z., Elaziz, M. A., Ewees, A. A., Ye, H. ve Jianhua, Z. (2019). Forecasting gold price fluctuations using improved multilayer perceptron neural network and whale optimization algorithm. Resources Policy, 61, 250-260.
  • Alkan, Ö., Genç, A., Oktay, E. ve Çelik, A. K. (2013). Electricity consumption analysis using spline regression models: the case of a Turkish province. Asian Social Science, 9, 231-240.
  • Alkan, Ö., Oktay, E., Genç, A. ve Çelik, A. K. (2017). An analysis of export-import coverage ratiosinTurkey using spline regressionmodels.Ekonomska Istraživanja / Economic Research, 30, 223-237.
  • Baur, D. G. ve McDermott, T. K. (2010). Is gold a safe haven? International evidence. Journal of Banking & Finance, 34(8), 1886-1898.
  • Boehmke, B. ve Greenwell, B. (2020). Hands-on machine learning with R (1 ed.): Chapman and Hall/CRC
  • Bouri, E., Jain, A., Biswal, P. C. ve Roubaud, D. (2017). Cointegration and nonlinear causality amongst gold, oil, and the Indian stock market: Evidence from implied volatility indices. Resources Policy, 52, 201-206.
  • Carmona, P., Climent, F. ve Momparler, A. (2019). Predicting failure in the U.S. banking sector: An extreme gradient boosting approach. International Review of Economics & Finance, 61, 304-323.
  • Carvalhal, A. ve Ribeiro, T. (2008). Do artificial neural networks provide better forecasts? Evidence from Latin American stock indexes. Latin American Business Review, 8(3), 92-110.
  • Chen, H.-H., Chen, M. ve Chiu, C.-C. (2016). The integration of artificial neural networks and text mining to forecast gold futures prices. Communications in Statistics - Simulation and Computation, 45(4), 1213-1225.
  • Chen,T. ve Guestrin,C.(2016). XGBoost: A scalable tree boosting system. Paper presented at the Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. Retrieved from https://doi.org/10.114 5/2939672.2939785
  • Chen,Y.,Lin,J.,Chen,Y. veWu,J.(2019). Financialforecastingwithmultivariate adaptive regression splines and queen genetic algorithm-support vector regression. IEEE Access, 7, 112931-112938.
  • Craven, P. ve Wahba, G. (1978). Smoothing noisy data with spline functions. Numerische Mathematik, 31(4), 377-403.
  • Değirmenci, N. ve Akay, A. (2017). Finansal verilerin ARIMA ve ARCH modelleriyle öngörüsü: Türkiye örneği. Eskişehir Osmangazi Üniversitesi İktisadi ve İdari Bilimler Dergisi, 12.
  • Eyduran, E., Akkus, O., Kazim, M., Tırınk, C. ve Tariq, M. (2017). Use of multivariate adaptive regression splines (Mars) in predicting body weight from body measurements in Mengali rams. Paper presented at the International Conference on Agriculture, Forest, Food Sciences and Technologies.
  • Eyduran, E. ve Duman, H. (2020). R yazılımı ile multivariate adaptive regression splines (MARS) uygulaması ders notları.
  • Fong-Ching, Y., Chao-Hui, L. ve Chaochang, C. (2020). Using market sentiment analysis and genetic algorithm-based least squares support vector regression to predict gold prices. International Journal of Computational Intelligence Systems, 13(1), 234-246.
  • Friedman, J. H. (1991). Multivariate adaptive regression splines. Ann. Statist., 19(1), 1- 67.
  • Friedman, J. H. (2001). Greedy function approximation: A gradient boosting machine. Ann. Statist., 29(5), 1189-1232.
  • Gangopadhyay, K., Jangir, A. ve Sensarma, R. (2016). Forecasting the price of gold: An error correction approach. IIMB Management Review, 28(1), 6-12.
  • Ghalayini, L. ve Farhat, S. (2020). Modeling and forecasting gold prices:Research Square. https://doi.org/10.21203/rs.3.rs-23825/v1 adresinden erişildi doi:10.21203/ rs.3.rs-23825/v1
  • Gulbe, A. ve Eyduran, E. (2019). ehaGoF: Calculates goodness of fit statistics (Version 0.1.0) [R Package].
  • Katris, C. (2020). Prediction of unemployment rates with time series and machine learning techniques. Computational Economics, 55(2), 673-706.
  • Keskin Benli, Y. ve Yıldız, A. (2014). Altın fiyatının zaman serisi yöntemleri ve yapay sinir ağları ile öngörüsü. Dumlupınar Üniversitesi Sosyal Bilimler Dergisi / Dumlupınar University Journal of Social Sciences, 213-224.
  • Khan, M. M. A. (2013). Forecasting of gold prices (Box Jenkins approach). International Journal of Emerging Technology and Advanced Engineering, 3(3), 662-670.
  • Kocatepe, C. İ. ve Yıldız, O. (2016). Ekonomik endeksler kullanılarak Türkiye’deki altın fiyatındaki değişim yönünün yapay sinir ağları ile tahmini. Düzce Üniversitesi Bilim ve Teknoloji Dergisi, 4, 926-934.
  • Kristjanpoller, W. ve 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.
  • Kuhn, M. (2020). caret: classification and regression training (Version 6.0-86) [R Package].
  • Livieris, I. E., Pintelas, E. ve Pintelas, P. (2020). A CNN–LSTM model for gold price time-series forecasting. Neural Computing and Applications.
  • Lu, C.-J., Lee, T.-S. ve Lian, C.-M. (2012). Sales forecasting for computer wholesalers: A comparison of multivariate adaptive regression splines and artificial neural networks. Decision Support Systems, 54(1), 584-596.
  • Ma, X., Sha, J., Wang, D., Yu, Y., Yang, Q. ve Niu, X. (2018). Study on a prediction of P2P network loan default based on the machine learning LightGBM and XGboost algorithms according to different high dimensional data cleaning. Electronic Commerce Research and Applications, 31, 24-39.
  • Miguéis, V. L., Camanho, A. ve Falcão e Cunha, J. (2013). Customer attrition in retailing: An application of multivariate adaptive regression splines. Expert Systems with Applications, 40(16), 6225-6232.
  • Milborrow, S. (2019). earth: multivariate adaptive regression splines (Version 5.1.2) [R package].
  • Mombeini, H. ve Yazdani Chamzini, A. (2015). Modeling gold price via artificial neural network. Journal of Economics, Business and Management, 3, 699-703.
  • Oktay, E., Genç, A. ve Alkan, Ö. (2012). İhracatın İthalatı Karşılama Oranlarının Parçalı Regresyonlarla Modellenmesi. Atatürk Üniversitesi Sosyal Bilimler Enstitüsü Dergisi, 16, 1-15.
  • Oktay, E.,Talas, E.,Alkan, Ö. ve Genç,A.(2012). Modeling with Linear SplineRegression of Turkish Tourism Demand. Journal of Selçuk University Natural and Applied Science, 1, 10-22.
  • Öndes, H. ve Oğuzlar, A. (2019). Yapay sinir ağlarıyla altın (TL/kg) fiyatı tahmini. Akademik Bakış Uluslararası Hakemli Sosyal Bilimler E-Dergisi(72), 249-262.
  • Pandey, A. C., Misra, S. ve Saxena, M. (2019, 8-10 Aug. 2019). Gold and diamond price prediction using enhanced ensemble learning. Paper presented at the 2019 Twelfth International Conference on Contemporary Computing (IC3).
  • Pierdzioch, C., Risse, M. ve Rohloff, S. (2016). A boosting approach to forecasting gold and silver returns: Economic and statistical forecast evaluation. Applied Economics Letters, 23(5), 347-352.
  • Risse, M. (2019). Combining wavelet decomposition with machine learning to forecast gold returns. International Journal of Forecasting, 35(2), 601-615.
  • Sami, I. ve Nazir, K. (2018). Predicting future gold rates using machine learning approach. International Journal of Advanced Computer Science and Applications, 8, 92- 99.
  • Sivalingam, K.C., Mahendran, S. ve Natarajan, S.(2016). Forecasting gold prices based on extreme learning machine. International Journal of Computers Communications & Control, 11(3).
  • Tianqi Chen, T. H. (2020). XGboost: extreme gradient moosting (Version 1.1.1.1) [R Package].
  • Verma, S., Thampi, G. T. ve Rao, M. (2020). ANN based method for improving gold price forecasting accuracy through modified gradient descent methods. IAES International Journal of Artificial Intelligence, 9(1), 46-57.
  • Visual XGBoost tuning with caret. (2020). Retrieved 08.07.2020, 2020, from https:// www.kaggle.com/pelkoja/visual-xgboost-tuning-with-caret
  • Wang, W., Shi, Y., Lyu, G. ve Deng, W. (2017). Electricity consumption prediction using XGBoost based on discrete wavelet transform. Paper presented at the 2nd International Conference on Artificial Intelligence and Engineering Applications (AIEA 2017).
  • Weng, F., Chen, Y., Wang, Z., Hou, M., Luo, J. ve Tian, Z. (2020). Gold price forecasting research based on an improved online extreme learning machine algorithm. Journal of Ambient Intelligence and Humanized Computing.
  • Yüksel, R. ve Akkoç, S. (2016). Altın fiyatlarının yapay sinir ağları ile tahmini ve bir uygulama. Doğuş Üniversitesi Dergisi, 17(1), 39-50.
  • Zhou, Y., Li, T., Shi, J. ve Qian, Z. (2019). A CEEMDAN and XGBOOST-based approach to forecast crude oil prices. Complexity, 2019, 4392785.
Year 2020, Issue: 83, 427 - 446, 30.09.2020

Abstract

References

  • Akin, M., Eyduran, S. P., Eyduran, E. ve Reed, B. M. (2020). Analysis of macro nutrient related growth responses using multivariate adaptive regression splines. Plant Cell, Tissue and Organ Culture (PCTOC), 140(3), 661-670.
  • Alameer, Z., Elaziz, M. A., Ewees, A. A., Ye, H. ve Jianhua, Z. (2019). Forecasting gold price fluctuations using improved multilayer perceptron neural network and whale optimization algorithm. Resources Policy, 61, 250-260.
  • Alkan, Ö., Genç, A., Oktay, E. ve Çelik, A. K. (2013). Electricity consumption analysis using spline regression models: the case of a Turkish province. Asian Social Science, 9, 231-240.
  • Alkan, Ö., Oktay, E., Genç, A. ve Çelik, A. K. (2017). An analysis of export-import coverage ratiosinTurkey using spline regressionmodels.Ekonomska Istraživanja / Economic Research, 30, 223-237.
  • Baur, D. G. ve McDermott, T. K. (2010). Is gold a safe haven? International evidence. Journal of Banking & Finance, 34(8), 1886-1898.
  • Boehmke, B. ve Greenwell, B. (2020). Hands-on machine learning with R (1 ed.): Chapman and Hall/CRC
  • Bouri, E., Jain, A., Biswal, P. C. ve Roubaud, D. (2017). Cointegration and nonlinear causality amongst gold, oil, and the Indian stock market: Evidence from implied volatility indices. Resources Policy, 52, 201-206.
  • Carmona, P., Climent, F. ve Momparler, A. (2019). Predicting failure in the U.S. banking sector: An extreme gradient boosting approach. International Review of Economics & Finance, 61, 304-323.
  • Carvalhal, A. ve Ribeiro, T. (2008). Do artificial neural networks provide better forecasts? Evidence from Latin American stock indexes. Latin American Business Review, 8(3), 92-110.
  • Chen, H.-H., Chen, M. ve Chiu, C.-C. (2016). The integration of artificial neural networks and text mining to forecast gold futures prices. Communications in Statistics - Simulation and Computation, 45(4), 1213-1225.
  • Chen,T. ve Guestrin,C.(2016). XGBoost: A scalable tree boosting system. Paper presented at the Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. Retrieved from https://doi.org/10.114 5/2939672.2939785
  • Chen,Y.,Lin,J.,Chen,Y. veWu,J.(2019). Financialforecastingwithmultivariate adaptive regression splines and queen genetic algorithm-support vector regression. IEEE Access, 7, 112931-112938.
  • Craven, P. ve Wahba, G. (1978). Smoothing noisy data with spline functions. Numerische Mathematik, 31(4), 377-403.
  • Değirmenci, N. ve Akay, A. (2017). Finansal verilerin ARIMA ve ARCH modelleriyle öngörüsü: Türkiye örneği. Eskişehir Osmangazi Üniversitesi İktisadi ve İdari Bilimler Dergisi, 12.
  • Eyduran, E., Akkus, O., Kazim, M., Tırınk, C. ve Tariq, M. (2017). Use of multivariate adaptive regression splines (Mars) in predicting body weight from body measurements in Mengali rams. Paper presented at the International Conference on Agriculture, Forest, Food Sciences and Technologies.
  • Eyduran, E. ve Duman, H. (2020). R yazılımı ile multivariate adaptive regression splines (MARS) uygulaması ders notları.
  • Fong-Ching, Y., Chao-Hui, L. ve Chaochang, C. (2020). Using market sentiment analysis and genetic algorithm-based least squares support vector regression to predict gold prices. International Journal of Computational Intelligence Systems, 13(1), 234-246.
  • Friedman, J. H. (1991). Multivariate adaptive regression splines. Ann. Statist., 19(1), 1- 67.
  • Friedman, J. H. (2001). Greedy function approximation: A gradient boosting machine. Ann. Statist., 29(5), 1189-1232.
  • Gangopadhyay, K., Jangir, A. ve Sensarma, R. (2016). Forecasting the price of gold: An error correction approach. IIMB Management Review, 28(1), 6-12.
  • Ghalayini, L. ve Farhat, S. (2020). Modeling and forecasting gold prices:Research Square. https://doi.org/10.21203/rs.3.rs-23825/v1 adresinden erişildi doi:10.21203/ rs.3.rs-23825/v1
  • Gulbe, A. ve Eyduran, E. (2019). ehaGoF: Calculates goodness of fit statistics (Version 0.1.0) [R Package].
  • Katris, C. (2020). Prediction of unemployment rates with time series and machine learning techniques. Computational Economics, 55(2), 673-706.
  • Keskin Benli, Y. ve Yıldız, A. (2014). Altın fiyatının zaman serisi yöntemleri ve yapay sinir ağları ile öngörüsü. Dumlupınar Üniversitesi Sosyal Bilimler Dergisi / Dumlupınar University Journal of Social Sciences, 213-224.
  • Khan, M. M. A. (2013). Forecasting of gold prices (Box Jenkins approach). International Journal of Emerging Technology and Advanced Engineering, 3(3), 662-670.
  • Kocatepe, C. İ. ve Yıldız, O. (2016). Ekonomik endeksler kullanılarak Türkiye’deki altın fiyatındaki değişim yönünün yapay sinir ağları ile tahmini. Düzce Üniversitesi Bilim ve Teknoloji Dergisi, 4, 926-934.
  • Kristjanpoller, W. ve 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.
  • Kuhn, M. (2020). caret: classification and regression training (Version 6.0-86) [R Package].
  • Livieris, I. E., Pintelas, E. ve Pintelas, P. (2020). A CNN–LSTM model for gold price time-series forecasting. Neural Computing and Applications.
  • Lu, C.-J., Lee, T.-S. ve Lian, C.-M. (2012). Sales forecasting for computer wholesalers: A comparison of multivariate adaptive regression splines and artificial neural networks. Decision Support Systems, 54(1), 584-596.
  • Ma, X., Sha, J., Wang, D., Yu, Y., Yang, Q. ve Niu, X. (2018). Study on a prediction of P2P network loan default based on the machine learning LightGBM and XGboost algorithms according to different high dimensional data cleaning. Electronic Commerce Research and Applications, 31, 24-39.
  • Miguéis, V. L., Camanho, A. ve Falcão e Cunha, J. (2013). Customer attrition in retailing: An application of multivariate adaptive regression splines. Expert Systems with Applications, 40(16), 6225-6232.
  • Milborrow, S. (2019). earth: multivariate adaptive regression splines (Version 5.1.2) [R package].
  • Mombeini, H. ve Yazdani Chamzini, A. (2015). Modeling gold price via artificial neural network. Journal of Economics, Business and Management, 3, 699-703.
  • Oktay, E., Genç, A. ve Alkan, Ö. (2012). İhracatın İthalatı Karşılama Oranlarının Parçalı Regresyonlarla Modellenmesi. Atatürk Üniversitesi Sosyal Bilimler Enstitüsü Dergisi, 16, 1-15.
  • Oktay, E.,Talas, E.,Alkan, Ö. ve Genç,A.(2012). Modeling with Linear SplineRegression of Turkish Tourism Demand. Journal of Selçuk University Natural and Applied Science, 1, 10-22.
  • Öndes, H. ve Oğuzlar, A. (2019). Yapay sinir ağlarıyla altın (TL/kg) fiyatı tahmini. Akademik Bakış Uluslararası Hakemli Sosyal Bilimler E-Dergisi(72), 249-262.
  • Pandey, A. C., Misra, S. ve Saxena, M. (2019, 8-10 Aug. 2019). Gold and diamond price prediction using enhanced ensemble learning. Paper presented at the 2019 Twelfth International Conference on Contemporary Computing (IC3).
  • Pierdzioch, C., Risse, M. ve Rohloff, S. (2016). A boosting approach to forecasting gold and silver returns: Economic and statistical forecast evaluation. Applied Economics Letters, 23(5), 347-352.
  • Risse, M. (2019). Combining wavelet decomposition with machine learning to forecast gold returns. International Journal of Forecasting, 35(2), 601-615.
  • Sami, I. ve Nazir, K. (2018). Predicting future gold rates using machine learning approach. International Journal of Advanced Computer Science and Applications, 8, 92- 99.
  • Sivalingam, K.C., Mahendran, S. ve Natarajan, S.(2016). Forecasting gold prices based on extreme learning machine. International Journal of Computers Communications & Control, 11(3).
  • Tianqi Chen, T. H. (2020). XGboost: extreme gradient moosting (Version 1.1.1.1) [R Package].
  • Verma, S., Thampi, G. T. ve Rao, M. (2020). ANN based method for improving gold price forecasting accuracy through modified gradient descent methods. IAES International Journal of Artificial Intelligence, 9(1), 46-57.
  • Visual XGBoost tuning with caret. (2020). Retrieved 08.07.2020, 2020, from https:// www.kaggle.com/pelkoja/visual-xgboost-tuning-with-caret
  • Wang, W., Shi, Y., Lyu, G. ve Deng, W. (2017). Electricity consumption prediction using XGBoost based on discrete wavelet transform. Paper presented at the 2nd International Conference on Artificial Intelligence and Engineering Applications (AIEA 2017).
  • Weng, F., Chen, Y., Wang, Z., Hou, M., Luo, J. ve Tian, Z. (2020). Gold price forecasting research based on an improved online extreme learning machine algorithm. Journal of Ambient Intelligence and Humanized Computing.
  • Yüksel, R. ve Akkoç, S. (2016). Altın fiyatlarının yapay sinir ağları ile tahmini ve bir uygulama. Doğuş Üniversitesi Dergisi, 17(1), 39-50.
  • Zhou, Y., Li, T., Shi, J. ve Qian, Z. (2019). A CEEMDAN and XGBOOST-based approach to forecast crude oil prices. Complexity, 2019, 4392785.
There are 49 citations in total.

Details

Primary Language Turkish
Journal Section Articles
Authors

Hayri Abar This is me

Publication Date September 30, 2020
Published in Issue Year 2020 Issue: 83

Cite

APA Abar, H. (2020). XGBOOST VE MARS YÖNTEMLERİYLE ALTIN FİYATLARININ KESTİRİMİ. EKEV Akademi Dergisi(83), 427-446.