林地变化驱动力研究中逻辑回归模型的应用

林地变化驱动力研究旨在揭示林地变化背后的驱动因素及其作用机制，是动态模拟、预测林地利用变化的前提工作。逻辑回归模型是研究林地变化驱动力的常见方法，但目前还没有学者对此进行系统的梳理。文中对逻辑回归模型在国内外林地变化驱动力研究中的应用进行综述，并提出对未来相关研究的建议; 通过文献分析法、归纳法等研究发现，相较于其他方法逻辑回归模型具有适用于因变量是分类变量的案例、重视空间机制等优势; 提出未来通过逻辑回归模型进行林地变化驱动力研究需要考虑数据的空间自相关性，选用恰当的研究单元、尺度以及变量。

Application of Logistic Regression Model in the Research on Driving Forces of Forestland Use Change

Forest land is the land for forest resource production and reproduction as the fundamental production elements in forestry activities. Its rational utilization is of great importance to manage and protect forest resources and conserve biodiversity. The study on the driving forces of forestland use change is to reveal the driving factors behind the changes and their operational mechanism, which is the precondition to dynamically simulate and forecast the changes in forestland use. Logical regression model is a common approach to studying the driving forces of forestland use change, but few scholars have made systematical review of this model. The paper reviewed the application of logical regression model to the national and international study on driving forces of forestland use change, and proposed suggestions for the future studies. The study, using literature review and inductive analysis, concluded that compared to other methods, logistic regression model has the advantages such as having the cases in which dependent variables are dummy, and more attention to spatial mechanism. And it was proposed that the future studies on forestland use change by logical regression model should focus on spatial autocorrelation of data, selection of appropriate research unit, scale and variables.