ABSTRACT
To forecast successfully financial time series data is not a trivial task. This type of data is characterized by internal complexity, high volatility and hidden stochastic dependencies. The factors that determines the values of the considered time series are unknown at future times and also must be forecasted. In this study, a univariate time series is modeled with data from bank deposits. A detailed analysis of the lagged variables allows to identify the predictor series to which a first-order trend and two temporal variables are added. Using the powerful machine learning algorithm Arcing from the class of gradient boosting methods the models for predicting of the examined data were built and analyzed. Aiming to avoid the presence of serial correlation in the residuals, an appropriate ARIMA error procedure is conducted to obtain the final hybrid Arcing-ARIMA models concerning the full sample and also for reduced 70% sample.
- Dadabada Pradeepkumar and Vadlamani Ravi. 2017. Forecasting financial time series volatility using Particle Swarm Optimization trained Quantile Regression Neural Network. Applied Soft Computing Journal 58, 35–52. https://doi.org/10.1016/j.asoc.2017.04.014Google ScholarCross Ref
- Hongju Yan and Hongbing Ouyang. 2018. Financial time series prediction based on deep learning. Wireless Personal Communications 102, 2, 683–700. https://doi.org/10.1007/s11277-017-5086-2Google ScholarDigital Library
- Dadabada Pradeepkumar and Vadlamani Ravi. 2017. Forex rate prediction: A hybrid approach using chaos theory and multivariate adaptive regression splines. In Proceedings of the 5th International Conference on Frontiers in Intelligent Computing: Theory and Applications (FICTA 2016). Advances in Intelligent Systems and Computing, Vol. 515, 219–227. Springer, Singapore. https://doi.org/10.1007/978-981-10-3153-3_22Google ScholarCross Ref
- Ioannis E. Livieris, Theodore Kotsilieris, Stavros Stavroyiannis, and Panagiotis E. Pintelas. 2020. Forecasting stock price index movement using a constrained deep neural network training algorithm. Intelligent Decision Technologies, 14, 3, 313–323. https://doi.org/10.3233/IDT-190035Google ScholarCross Ref
- Ritika Singh and Shashi Srivastava. 2017. Stock prediction using deep learning. Multimedia Tools and Applications 76, 18, 18569–18584. https://doi.org/10.1007/s11042-016-4159-7Google ScholarDigital Library
- Zhigang Yang, Yi Yang, Durong Yin, Mao Yang, and Lian Li. 2019. Research on data analysis for time deposit of bank customers based on ensemble learning. In Proceedings of the 2019 IEEE 3rd International Conference on Electronic Information Technology and Computer Engineering (EITCE 2019). 9094858, 1325–1330. https://doi.org/10.1109/EITCE47263.2019.9094858Google ScholarCross Ref
- Christopher Krauss, Xuan Anh Do, and Nicolas Huck. 2017. Deep neural networks, gradient-boosted trees, random forests: Statistical arbitrage on the S&P 500. European Journal of Operational Research 259, 2, 689–702. https://doi.org/10.1016/j.ejor.2016.10.031Google ScholarCross Ref
- Nesreen K. Ahmed, Amir F. Atiya, Neamat El Gayar, and Hisham El-Shishiny. 2010. An empirical comparison of machine learning models for time series forecasting. Econometric Reviews 29, 5-6, 594–621. https://doi.org/10.1080/07474938.2010.481556Google ScholarCross Ref
- Jong-Min Kim, Dong H. Kim, and Hojin Jung. 2021. Applications of machine learning for corporate bond yield spread forecasting. North American Journal of Economics and Finance 58, Art. No 101540. https://doi.org/10.1016/j.najef.2021.101540Google ScholarCross Ref
- Paravee Maneejuk and Wilawan Srichaikul. 2021. Forecasting foreign exchange markets: further evidence using machine learning models. Soft Computing 25, 12, 7887–7898. https://doi.org/10.1007/s00500-021-05830-1Google ScholarDigital Library
- Anastasios Petropoulos, Vasilis Siakoulis, Evangelos Stavroulakis, and Nikolaos E. Vlachogiannakis. 2020. Predicting bank insolvencies using machine learning techniques. International Journal of Forecasting 36, 3, 1092–1113. https://doi.org/10.1016/j.ijforecast.2019.11.005Google ScholarCross Ref
- Michalis Doumpos, Constantin Zopounidis, Dimitrios Gounopoulos, Emmanouil Platanakis, and Wenke Zhang. 2023. Operational research and artificial intelligence methods in banking. European Journal of Operational Research 306, 1, 1–16. http://dx.doi.org/10.2139/ssrn.4085812Google ScholarCross Ref
- Ibomoiye D. Mienye and Yanxia Sun. 2022. A survey of ensemble learning: Concepts, algorithms, applications, and prospects. IEEE Access 10, 99129–99149. http://dx.doi.org/10.1109/ACCESS.2022.3207287Google ScholarCross Ref
- Christo El Morr, Manar Jammal, Hossam Ali-Hassan, and Walid El-Hallak. 2022. Overview of machine learning algorithms. In Machine Learning for Practical Decision Making. International Series in Operations Research and Management Science, Vol. 334, 61–115. Springer, Cham. https://doi.org/10.1007/978-3-031-16990-8_3Google ScholarCross Ref
- Wassima Lakhchini, Rachid Wahabi, and Mounime El Kabbouri. 2022. Artificial Intelligence & Machine Learning in Finance: A literature review. International Journal of Accounting, Finance, Auditing, Management and Economics 3, 6-1, 437–455. https://doi.org/10.5281/zenodo.7454232Google ScholarCross Ref
- Leo Breiman. 1998. Arcing classifiers. The Annals of Statistics 26, 3, 801–849.Google ScholarCross Ref
- IBM SPSS Software. https://www.ibm.com/analytics/spss-statistics-softwareGoogle Scholar
- SPM (Salford Predictive Modeler), Machine Learning and Predictive Analytics Software. https://www.minitab.com/en-us/products/spm/Google Scholar
- Lina Yordanova, Gabriela Kiryakova, Petya Veleva, Nadezhda Angelova, and Antoaneta Yordanova. 2021. Criteria for selection of statistical data processing software. In Proceedings of the IOP Conference Series: Materials Science and Engineering 1031, 1, 012067. https://doi.org/10.1088/1757-899X/1031/1/012067Google ScholarCross Ref
- Bulgarian National Bank. www.bnb.bgGoogle Scholar
- George E.P. Box, Gwilym M. Jenkins, and Gregory S. Reinsel. 1994. Time Series Analysis, Forecasting and Control (3rd. ed.). Prentice-Hall, New Jersey.Google Scholar
- Eric Bauer and Ron Kohavi. 1999. An empirical comparison of voting classification algorithms: bagging, boosting, and variants. Machine Learning 36, 105–139. https://doi.org/10.1023/A:1007515423169Google ScholarDigital Library
- In‐Kwon Yeo and Richard A. Johnson. 2000. A new family of power transformations to improve normality or symmetry. Biometrika 87, 4, 954–959. https://doi.org/10.1093/biomet/87.4.954Google ScholarCross Ref
- Ivaylo V. Boyoukliev, Snezhana G. Gocheva-Ilieva, and Hristina N. Kulina. 2022. Time series modeling and forecasting of deposits in foreign currency using CART ensemble and bagging. In Proceedings of the 13th International Hybrid Conference for Promoting the Application of Mathematics in Technical and Natural Sciences (AMiTaNS’21). AIP Conference Proceedings 2522, 050003, 12 pages. https://doi.org/10.1063/5.0101185Google ScholarCross Ref
Index Terms
- Forecasting Volatility of Bank Deposits of Individuals Using Hybrid Arcing -ARIMA Approach: Forecasting Volatility of Bank Deposits
Recommendations
Forecasting volatility of crude oil futures using a GARCH–RNN hybrid approach
AbstractVolatility is an important element for various financial instruments owing to its ability to measure the risk and reward value of a given financial asset. Owing to its importance, forecasting volatility has become a critical task in financial ...
Effective options trading strategies based on volatility forecasting recruiting investor sentiment
This study investigates an algorithm for an effective option trading strategy based on superior volatility forecasts using actual option price data for the Taiwan stock market. The forecast evaluation supports the significant incremental explanatory ...
Hybrid Model Using Interacted-ARIMA and ANN Models for Efficient Forecasting
Multi-disciplinary Trends in Artificial IntelligenceAbstractWhen two models applied to the same dataset produce two different sets of forecasts, it is a good practice to combine the forecasts rather than using the better one and discarding the other. Alternatively, the models can also be combined to have a ...
Comments