Session 1: Tues March 7 | Session 2: Friday March 10
Session 3: Tues March 14 | Session 4: Friday March 17
Overview
This webinar series “An Introduction to Spatial Regression Models” provides an overview of spatial regression models and their application in the analysis of spatially-referenced data. Taught through a lens of health, this short course introduces users to the spatial-analytic framework and how spatial regression can be used to explore the importance of spatial relationships between health, social and environmental processes. The role of spatial autocorrelation in spatial data is a central focus of this course. We will discuss how it arises, how it is measured, and its treatment as either a nuisance or an interesting phenomenon using spatial regression techniques.
This course is intended for users with basic knowledge in linear regression and who are interested in learning how to incorporate and adjust for spatial complexities and context. It therefore focuses on the application of models, when and how to run and interpret the results with general formulas presented, but does not delve into their statistical theory. Users will be introduced to the open-source software GeoDa as well as R, a free, open-source language and environment for statistical computing and graphics. No prior working knowledge of GeoDa or R is required, but some familiarity with R is recommended.
The webinar series will be divided into four 2-hour sessions. Background theory and live demos will be presented during the live session, with take-home practice assignments and readings provided to further explore the theory, methods and applications of spatial regression.
Prior required knowledge
Familiarity with R is an asset though not mandatory for enrollment. Knowledge in basic methods for descriptive statistics, statistical inference, and linear regression will be assumed.
Webinar objectives
By the end of this webinar series, participants will be able to:
- Assess global and local spatial autocorrelation using GeoDa and R
- Conduct exploratory spatial data analysis (ESDA) using GeoDa and R
- Run and interpret spatial error and spatial lag regression models using GeoDa and R
- Run and interpret Geographical Weighted Regression (GWR) models using R
Webinar Format
The interactive webinar software will provide remote access for students to view the instructor's screen, listen to the lecture in real time, and ask questions. The instructor will provide lecture slides (PowerPoint) and required readings prior to the start of the webinar. For practice between webinar sessions and for follow up study, students will also receive training data and programming code for GeoDa and R.
Instructor biography
Anders Erickson, PhD, is a Post-doctoral Fellow at the University of British Columbia, School of Population and Public Health. He completed an Interdisciplinary PhD between the Division of Medical Sciences and Department of Geography at the University of Victoria. His research interests include environmental, chronic disease and perinatal epidemiology, with a particular emphasis on the interactions between the social, physical and biological environments and the statistical methods used for analyzing such high dimensional data.
Workshop fees
- Regular rate: $265
- Student Rate: $165
Course content
Session 1: Introduction to Spatial Analytic Framework
- What is a spatial question?
- Why do I need spatial regression? Why is it important?
- Exploratory spatial data analysis (ESDA) mapping and visualizations using GeoDa
Session 2: Spatial Autocorrelation
- Spatial autocorrelation overview – how does it arise?
- Measuring spatial autocorrelation
- The spatial weight matrix: what is it, why is it important, and how to choose one?
Session 3: Spatial Regression Models
- What is spatial regression and how is it different from OLS?
- What is a spatial error model?
- What is a spatial lag model?
- How do error and lag models differ and when do I use them?
Session 4: Spatial Non Stationarity & Local Regression Models
- What is spatial non-stationarity and why is it important?
- What is the difference between global and local spatial regression models?
- Application of local (Geographical Weighted Regression, GWR) models
