Introduction to spatial analysis in R

20-21 May 2019, Bilbao,
Universidad del País Vasco

Seminar held by:
Tomas Formanek
Department of Econometrics (Faculty of Informatics and Statistics)
University of Economics, Prague

Day 1 (20. 5. 2019): R/RStudio for spatial analysis

Seminar 1: R & RStudio (09:00 - 10:15)

• General R / RStudio introduction
• Global environment, commands, packages
• Linear regression in R (syntax)

Seminar 2: Data handling and plotting I. – Tools (10:30 - 11:45)

• Data types and basic data operations (filtering and merging data frames)
• {tidyverse} packages, introduction
• Simple plots with {ggplot2}, adding features (layers)
• Data handling with focus on long and short format with {tidyr} package
• {dplyr} package and its basic features

Seminar 3: Data handling and plotting II. – Choropleths (12:00 - 13:15)

• {sf} package for handling spatial data
• Eurostat data and R (data search, download)
• Maps in {ggplot2}
• Choropleths (infomaps) in {ggplot2}

Day 2 (21. 5. 2019): Spatial analysis – an introduction

Seminar 1: Introduction to spatial statistics with applications in R (09:00 - 10:15)

• Spatial data: basic features and analysis tools (stationarity, semivariogram)
• Simple predictions based on spatial data (spatial interpolation, krigging)

Seminar 2: Introduction to spatial econometrics (10:30 - 11:45)

• Spatial structure (definition of neighbors)
• Spatial dependency and corresponding tests

Seminar 3: Spatial econometric models (12:00 - 13:15)

• Spatial econometric models for cross-sectional (CS) data
• Estimation and interpretation of selected spatial CS model types
• Robustness of the estimated models with respect to changing spatial structure

Instructions for course participants

Please observe the following instructions:

• Bring your own laptop/notebook, make sure you have access to the Internet (e.g. through eduroam at Universidad del País Vasco)
• Have R version 3.6.0 (3.5.3 is also OK) installed on your device (PC/Mac/Linux)
• Have RStudio installed (free desktop version 1.2.1335 or newer)
  • If you experience problems while compiling Rmd files, you may consider installing the newest version of Pandoc document converter.
  • (Pandoc is part of RStudio instalation, new version solves the file conversion problems)
• Setup an R working directory (e.g. on your machine’s Desktop) - see instructions Changing the Working Directory
• Download all files from GitHub repository to your R working directory (if not familiar with Git/GitHub, just use the green button “Clone or download”)
• Install all packages for the course - run the file R_Packages.R downloaded from GitHub repository for this course.
• If you have no previous exprerience with R/RStudio and have troubles following the above instructions, you may consider taking a free R/RStudio online course before attending to the seminar on spatial analysis, e.g.: R Basics - R Programming Language Introduction

Additional study materials and resources

• Kleiber, Zeileis (2008): Applied Econometrics with R
  • From introductory to advanced econometric analysis using R/RStudio
• Bivand, Pebesma, Gómez-Rubio (2013): Applied Spatial Data Analysis with R
  • Predates {sf} package
• Simple Features for R
• List of R packages for different types of spatial analysis

Home page

https://formanektomas.github.io/SpatialAnalysisBilbao2019/