摘要
研究目的:探索改进时间序列预测法与多因素预测法在区域耕地资源变动预测及驱动机制分析方面局限性的途径。研究方法:(1)引入协整理论有关方法对时间序列数据平稳性及协整性进行分析和评估,为驱动机制研究提供思路;(2)探索构建具有"影响滞后"机制的前移回归模型预测方法。研究结果:在对福建省的实证分析案例中,上述方法运用效果较好,模型输出的预测数据及所作的驱动机制分析可为该省今后的种植业生产规划、农业土地规划与决策提供参考。研究结论:协整理论与前移回归方法的有机结合,可为区域耕地资源变动预测及驱动机制研究提供一种有效的工具。
The purpose of this paper is to explore how to release the restrictions of time series method and multi-factor method used in the prediction of regional farmland change and the analysis on its driving mechanism. Method employed first is the co-integration theory, which is used to analyze and assess the stability and co-integration degree of time series data in order to explore further the driving mechanism. And the second method is to explore and establish the prediction methods based on forward regression model involving ＂influence-lag＂ effect. The results indicate： （1） performance of the methods mentioned above are good according to the empirical analysis case of Fujian Province; （2） the predicted data from the models and the analyzed driving mechanism can provide reference points for agricultural production planning, land use planning and other policy designs. It is concluded that the combination of co-integration theory and forward regression model can be an effective tool for the prediction of regional farmland resource change and exploring the relevant driving mechanism.
出处
《中国土地科学》
CSSCI
北大核心
2010年第11期10-14,9,共6页
China Land Science
基金
国家社会科学基金项目(03BTJ004)
关键词
土地利用
耕地资源
预测
驱动机制
协整性
前移回归
福建省
land use
farmland resources
prediction
driving mechanism
co-integration
forward regression
Fujian Province
