ANALYSIS OF THE SUPPORT VECTOR MACHINE ALGORITHM IN COMPARISON TO TRADITIONAL MARKET MOVEMENTS PREDICTION METHODS
DOI:
https://doi.org/10.26906/SUNZ.2024.3.089Keywords:
stock markets, neural networks, genetic algorithm, economy, artificial intelligence, market decisions supportAbstract
Topicality. The development and adjustment of a universal algorithm for making market decisions is the primary task of experts and investors of stock markets around the world. Modern technologies offer relevant solutions using neural networks and artificial intelligence. However, existing solutions are only suitable for limited use and require large amounts of training data. The goal of this work is analysis of the support vector machine performance on market decisions adjustment, and its comparison with traditional strategies and methods of analysis of market movements. The object of research is the process of making market decisions based on the neural networks. The subject of research is the support vector machine algorithm. Results. In this paper, was analyzed the support vector machine algorithm on market models for various approaches and modifications, also were offered ways to improve the efficiency of the decisions made. Conclusions. The support vector machine has demonstrated greater efficiency and reliability than classical methods of analysis, on models with high market volatility. This algorithm shows positive results in crisis and unstable market models, its use is optimal for capital preservation.Downloads
References
Iacomin, R. (2015), “Stock market prediction”, 2015 19th International Conference on System Theory, Control, and Computing
Prasad Das (2012). Support Vector Machines at Prediction of Futures Costs in Indian Stock Market. International Journal of Computer Applications.
F.E.H. Tay et al. “Application of support vector machines in financial time series forecasting”, Omega (2001)
Karathanasopoulos (2016). “Stock market anticipation utilizing evolutionary support vector machines: an application of the ASE20 index”, The European Journal of Finance
Yaloveha V., Podorozhniak A., Kuchuk H. Convolutional neural network hyperparameter optimization applied to land cover classification. Radioelectronic and computer systems. 2022. No. 1. C. 115–128. DOI: https://doi.org/10.32620/reks.2022.1.09
Rosillo (2014), “Stock market simulation utilizing support vector machines. Journal of Forecasting”
M.T. Leung et al. “Forecasting stock indicesa comparison of classification and level estimation models”, International Journal of Forecasting (2000)
Y.S. Abu-Mostafa et al. “Introduction to financial forecasting”, Applied Intelligence (1996)
Studies, E. (2000). A Stock Market Prediction Method Based on Support Vector Machines and Independent Component Analysis
Yetis, Y., Kaplan, H., & Jamshidi, M. (2014), “Stock market prediction by using the artificial neural network”, 2014 World Automation Congress
Ichinose, K., & Shimada, K. (2016), “Stock market prediction from news on the web and a new evaluation approach in trading. Proceedings” 2016 5th IIAI International Congress on Advanced Applied Informatics
