The stationary fuzzy stochastic time series prediction methods based on fuzzy stochastic process

In the treatment of the practical problems, people often encounter some fuzzy random phenomenon, which cannot be effectively described and processed with classic probability theory methods. In recent years, a large number of techniques have been proposed for predicting the stationary fuzzy stochastic time series. The classic stochastic time prediction method is one kind of prediction method that considers the change regularity of the variables over time and establishes a mathematical model using the previous statistical data of the variable. Because it just needs the historical data of the sequence itself, so this kind of method is widely used. In this paper, we have presented one kind of prediction method of the stationary fuzzy stochastic time series, the research of one kind of prediction approach for the fuzzy stochastic time series is made. This paper has analyzed the characteristics of the prediction approach and established several kinds of linear fuzzy random models. Also we have made the recognition and parameter estimation of the stationary fuzzy random time series.