基于新维无偏灰色马尔可夫的农产品产量预测模型

The Frediction Model to Agriculture Products Yield Based on New Information Grey Markov

针对农产品产量随机性和波动性的特点,建立了新维无偏灰色马尔可夫的预测模型。该模型是在传统灰色马尔可夫理论的基础上,对马尔科夫模型进行了改进,提高了预测精度。通过无偏灰色模型预测变化趋势,借助马尔可夫模型处理随机性波动,同时在每一步的预测中,用新信息代替旧信息,对原始数据进行等维处理,更新建模数据。以人均粮食产量为例进行仿真实验,平均相对误差达到0.25%,且预测误差的变化波动性减小。结果表明,提出的模型有较好的预测精度,能够满足农产品产量预测精度的要求,适合中长期预测。

The agriculture products yield has the characteristics of random and fluctuation, so this paper proposed a prediction model to agriculture products yield based on grey theory. Based on the traditional grey forecasting model and Markov chain theory, equal dimension and new information unbiased grey Markov forecasting model was established. The development tendency is imitated by the unbiased Grey model, and the stochastic volatility is dealt with by the Markov model. The newest data are gradually added while the oldest one is removed from original data sequence. The simulation experiment was carried out with food production per capita as an example, and the average relative error is 0.25%. Case study showed that the model has fewer errors and better forecasting precision, especially for medium and long - term prediction.