Incorporating the Markov chain concept into fuzzy stochastic prediction of stock indexes

In this paper we incorporate the Markov chain concept into fuzzy stochastic prediction of stock indexes in order to attain better accuracy and confidence. By the fuzzy stochastic method, parameters for prediction are produced using a fuzzy linguistic summary, whereas by our proposed model, parameters are determined using both a fuzzy linguistic summary and the probabilities of stock indexes rising or falling. This model, whose performance having been tracked for three months, has proved to be significantly better for stock index prediction.