PROFITABILITY ANALYSIS OF ARTIFICIAL INTELLIGENCE METHODS AND TECHNICAL INDICATORS IN THE FOREX EXCHANGE

摘要

For algorithmic trading many methods have been suggested. This article will be devoted to predicting the behavior of currency pairs on the Forex exchange. Several methods of artificial intelligence and statistical analysis will be used to predict the future of forex currency pairs. We adopt the idea of transforming the window of time series into the spectral plane, and then analyzing the clusters using hierarchical clustering, Kohonen maps and evolutionary algorithms. Next, we will use a radically different approach to time series analysis in the time domain: fuzzy time series and genetic algorithm. In the numerical study we will compare the success of artificial intelligence algorithms and multivariate statistical analysis with the success of technical indicators MACD and ROC.