| Grey system model and Markov chain can be applied to the time series forecasting, but their own characteristics and limitations. Gray-Markov chain model decompose random time sequence for gray model to forecast the trend of changes in the sequence and random changes in the Markov chain two parts, play their respective advantages and overcome the defects of both, improve the prediction accuracy.This paper mainly studies the Improvement issues of Gray Markov chain model, to eliminate the grey bias and improve anti-jamming performance of the standard Grey-Markov forecasting model, the unbiased grey theory and fuzzy classification are introduced into the Grey-Markov forecasting model and a new method named the Unbiased Grey-Fuzzy-Markov Chain Method is proposed. This method uses the Unbiased Gray model eliminate gray bias, and uses fuzzy mathematical theory on fuzzy division of the state, to better solve the deviation of gray Markov model. Finally, we apply the model to the problem of the per capita GDP of Gansu Province, select the per capita GDP data for1990-2011years of Gansu Province to simulate, predicted the trend of per capita GDP of Gansu Province in the next few years, and do a dynamic analysis for it, and elaborated the Unbiased Grey-Fuzzy-Markov Chain forecasting methods in specific applications, draw a realistic forecast results, the accuracy and reasonableness of the result is significantly better than the traditional Gray-Markov model. |
