Algoritmické obchodování

Abstract

The field of quantitative mathematics and related algorithmic trading has become a common practice for beginning serial investors and a necessary tool for experienced day traders. Algorithmic trading combines elements of psychology, economics, mathematics and computer science. The aim of this bachelor thesis is to explore the possibilities of using neural networks to predict market trends. The thesis first introduces possible software tools and key areas of algorithmic trading for stock market data analysis and prediction. Real data from Binance exchange was used as the exchange data, which is analyzed and modified for neural networks to better predict the market. The target feature to predict and technical indicators were selected. The aim is to use all these elements as additional knowledge to better estimate the market behavior. These elements are used in two strategies - conservative and greedy. The two strategies are compared and evaluated against each other, both in terms of final returns and success rates. The conservative strategy was found to have a better return and the greedy strategy had a better success rate.