When using Grey Markov Model to predict the stock price, the state space is divided by an absolute set theory, so the result's accuracy is not high, and even error. In order to improve the accuracy of predictive value, this paper proposes a model based on fuzzy gray prediction and Markov theory. After mastering the trend of the raw data through the Gray Model, we can divide the state space into fuzzy partitions, then combine with the results of the relative residual test method, effectively reduce the errors caused by the traditional absolute division. Experimental results show that compared with the Gray Markov model, the accuracy of predictive value has increased significantly in a short period, and it can be applied to the prediction of stock market.
