Predição do mercado de ações usando Hidden Markov Model

Abstract

Stock markets are complex systems due its non-stationarity, the parameters are always in constant changes like economic conditions and changes in the politics of companies. There are several researches in predicting the stocks with Artificial Intelligence and Machine Learning techniques, such as Artificial Neural Networks, Support Vector Machine, Fuzzy Logic, Pattern Recognition, which, in the latter, the Hidden Markov Model, focus of this paper, fits. Because it is a stochastic model, where events are random, the results show that HMM can be well used in predicting the behavior of stocks based on their historical performance.