基于灰色理论—马尔可夫模型对畜禽粪便量增长的预测

Prediction of the Amount of Livestock and Poultry Feces by Gray-Markov Models

结合灰色GM（1，1）模型和马尔可夫模型构建一种合理的灰色马尔可夫预测模型。按相对值的方法进行状态划分，先用灰色GM（1，1）模型预测，再用马尔可夫模型对预测结果进行优化，并将由此建立的灰色马尔可夫模型运用于1997～2005年衡阳地区畜禽粪便数据来预测2006年畜禽粪便量。结果表明，这种模型的预测精度高达98．4554％，明显优于单独使用GM（1，1）模型的预测结果。

Combining GM（1,1） model and Markov model, a reasonable method is presented in this paper. The novel method was established to predict livestock and poultry feces of Hengyang region based on Gray and Markov model to Gray Markove mod el. According to ways of the relative quantity to divide state, GM（1,1） model was firstly used to optimize the data, then, Markov model was used to optimize the results from the GM（1,1） model again. Using the livestock and poultry data of the years from 1997 to 2005 the amount of livestock and poultry is predicted. The result showed that the precision of the model got to 98.46 %, and it was superior to the model of GM （1,1）.