课程1：Statistical Learning with Spatial Data空间数据统计学习

主讲老师：李斌 美国中密歇根大学地理系教授，曾任中密歇根大学地理系和地理信息科学中心主任，现任中密歇根大学科学技术学院中国事务院长助理、武汉大学讲座教授和国际地学计算联合中心副主任、中密歇根大学-江西师大中美大湖流域联合研究中心执行主任、华南师大特聘教授。

助教：龚君芳18971626856

时间：2019年5月20日-5月24日（23日下午，其他日期上午）

5月20日上午9：00-12：00

5月21日上午9：00-12：00

5月22日上午9：00-12：00

5月23日下午14：30-17：30

5月24日上午9：00-12：00

地点：信息楼201

要求：自带笔记本电脑，安装R（https://www.r-project.org/）和ESF Tool（https://thesaar.github.io/）

课程简介：

This course introduces the methods and techniques for performing regression modeling with geographically reference data. The first part of the course reviews linear regression and Generalized Linear Regression, through which the importance of geographic effects is illustrated. The second part introduces implementations of spatial models for both linear models and GLM, which focuses on the Spatial Autoregressive Models and Eigenvector Spatial Filtering. The course concludes with real world applications that demonstrate concepts and procedures of conducting spatial regression modeling. The class will be taught with a combination of lecture and real time demonstrations. Software R and ESF Tool will be used for labs and demonstration. Students are expected to have backgrounds in linear algebra and introductory statistics.

Below is the list of the tentative topics.

• Day 1 Introduction to statistical learning; exploratory data analysis; linear regression

• Day 2 Generalized Linear Regression

• Day 3 Spatial autoregressive models

• Day 4 Eigenvector spatial filtering

• Day 5 Applications

The following books are used as references.

• James, et al., An Introduction to Statistical Learning, Springer. http://www-bcf.usc.edu/~gareth/ISL/

• Smith and Goodchild, Geospatial Analysis. https://www.spatialanalysisonline.com/HTML/index.html

• Chun and Griffith, 2013. Spatial Statistics & Geostatistics, Sage.