基于改进灰色马尔可夫模型的花生产量预测

Application of improved Grey-Markov model to peanut yield prediction

【目的】为了提高花生产量的中长期预测精度,提出一种改进的灰色马尔可夫模型。【方法】首先,以中国2010—2020年花生产量数据为研究样本,在传统灰色模型的基础上进行加权滑动平均处理,建立滑动灰色模型;然后,利用马尔可夫链对预测结果进行修正,得到滑动灰色马尔可夫模型;最后,采用新陈代谢的思想,对原始数据序列做等维新息处理,构建新维滑动灰色马尔可夫模型。【结果】新维滑动灰色马尔可夫模型的平均相对误差比滑动灰色模型和滑动灰色马尔可夫模型分别降低了80.00%和48.89%,并预测出未来5年中国花生产量将以2%左右的增长率增长。【结论】本研究结果可为其他农作物产量预测提供一种科学合理的思路。

[Objective]An improved Grey-Markov model was proposed to improve the medium and long-term prediction accuracy of peanut yield in China.[Method]Firstly,taking the data of peanut yield from 2010 to 2020 in China as the research sample,the weighted moving average was processed to establish the sliding grey model on the basis of the traditional grey model;then,the prediction results were modified by the Markov chain to obtain the sliding Grey-Markov model;finally,the idea of metabolism was adopted to process the equal dimension and new information of the original data sequence,constructing a new dimensional sliding Grey-Markov model.[Result]The mean relative error of the new dimensional sliding Grey-Markov model is 80.00%and 48.89%lower than that of the sliding grey model and the sliding Grey-Markov model,respectively.At the same time,it is predicted that the peanut yield of China will increase by about 2%in the next five years.[Conclusion]The results can provide a scientific and reasonable approach for prediction of other crop yields.