XIE Yong , LIU Ming,  ZHANG Yilai

(Jingdezhen Ceramic Institute, Jingdezhen, Jiangxi 333403)

Abstract: Gray model is one of the most effective ways in the economic prediction methods, but gray model is poor at predicting larger volatility sequences. An improved gray-Markov model was presented in this paper, which ran ln (x) function for the original sequence to improve data smoothness and used the minimum distance method to reduce the relative error of the gray model, and used the Markov chain to correct the predicted results. The improved model was applied to predict the Jingdezhen Ceramic Industrial output value, and compared with the traditional gray and the gray-Markov models. The results showed that the improved model prediction's relative error was much smaller than that of the gray model, its prediction accuracy was better than that of the Grey-Markov model, and it was more responsive to economic trends.

Key words:  grey theory; Markov chain; relative error; prediction; grey-Markov model