Fisher算法及加权马尔可夫链模型的耕地需求量预测方法——以山东省聊城市为例  

Prediction of cultivated land demand based on Fisher arithmetic and Markov Chain Model with Weights——a case study of LiaoCheng city

作　　者：聂芹[1] 

机构地区：[1]聊城大学环境与规划学院,山东聊城252059

出　　处：《测绘科学》2011年第1期29-31,共3页Science of Surveying and Mapping

摘　　要：根据聊城市1949-2007年的耕地数据,以耕地变化率来描述耕地数量变化的速率,借助Fisher最优分割法将耕地变化率状态分为5个等级,利用规范化后的各阶自相关系数为权重,运用加权马尔可夫链模型对聊城市耕地需求量进行预测。结果表明,预测值与实际值的相对误差为0.01%,预测结果非常理想,同时也验证了加权后的马尔可夫预测法更具科学性和实用性。Cultivated Land in Liaocheng city was predicted based on Markov Chain Model with Weights. The Cultivated Land quantitative change was described by the change rate of Cultivated Land based on Cultivated Land from 1949 to 2007 in Liaocheng city. By adopting Fisher arithmetic, the rate of change was divided into five grades. Standardized self-correlative coefficients based on the special characteristics of correlation among the historical stochastic variables were regarded as weights. The experimental results showed that the predicted result was satisfying and the model with weights was more scientific and practical

关 键 词：耕地需求量 Fisher最优分割法 加权马尔可夫链模型 

分 类 号：P934[天文地球—自然地理学]