数字地价模型的构建方法及参数选择

发布时间：2018-03-09 17:06

  本文选题：数字地价模型　切入点：地统计学　出处：《统计与决策》2017年20期 　论文类型：期刊论文

【摘要】：基于地统计内插构建数字地价模型是目前国内研究地价空间分布的常用手段。既往的研究大多侧重空间变异分析或不同模型结果的比较,较少探讨一个完整的建模过程及过程中一些重要参数如何科学选择与设置。文章以2008—2014年石家庄市主城区土地交易数据为例,借助ArcGIS、GS+软件平台,深入探讨了基于地统计学原理,利用Kriging插值的方法,构建数字地价模型的完整过程,并重点探究此建模过程中获取最优参数的方法。结果表明:基于地统计方法构建的数字地价模型,能够合理地模拟区域地价空间分布形态;并通过探索性空间数据分析、变异函数分析、不同模型拟合优度对比、交叉验证精度检验等手段,有效地保证了建模参数的科学性和准确性。
[Abstract]:To construct digital land price model based on geostatistics interpolation is a commonly used method to study the spatial distribution of land price in China at present. Most previous studies have focused on spatial variation analysis or comparison of the results of different models. This paper takes the land transaction data of Shijiazhuang City from 2008 to 2014 as an example, and discusses the principle of geostatistics based on ArcGISGS software platform. Using the Kriging interpolation method, the complete process of constructing the digital land price model is discussed, and the method of obtaining the optimal parameters in the process of the model is studied. The results show that the digital land price model based on the geostatistics method is constructed. The spatial distribution of regional land price can be simulated reasonably, and by means of exploratory spatial data analysis, variance function analysis, comparison of goodness of fit of different models, cross-validation accuracy test, etc. The scientific and accuracy of modeling parameters are ensured effectively.
【作者单位】： 张家口学院理学系;河北师范大学资源与环境科学学院;
【基金】：河北省高等学校科学技术研究项目(QN2017522)
【分类号】：F224;F299.23
，

本文编号：1589450