Welcome the GIS for Economists - Crash Course -. The aim of the course is to provide you with a basic knowledge to master the spatial data analysis toolbox. It offers alternative solutions to a broad range of empirical problems of applied economics.

We introduce students to the basic tools to analyze, create and manipulate different types of datasets from Geographic Information Systems (GIS). The aim of the course is to provide students with a basic knowledge to master the spatial data analysis toolbox, which offers alternative solutions to a broad range of empirical problems of applied economics. It is a practical course in which students are required to have an active participation through practices during lectures and problem sets. We will explore the basic GIS Desktop tools (i.e. ArcGIS and QGIS) as well as some GIS applications in standard statistic packages (R). However, an especial emphasis will be placed on building standalone solution based on Python (e.g. GDAL, FIONA, among others).

In this website, you will find slides and additional materials to encourage you to go further when including GIS tools in your research.

Before starting, I would like you to ask you installing these tree programs: (i) QGIS, and (2) R- RStudio. We will need them during the course.

Installing QGIS

QGIS will be the GIS Desktop tool that will be use during the course. It is a free and open source software. It is a very cool project that brings all the power of GIS analysis for everybody. We will discuss its pro and cons over other GIS Desktop tools (e.g. ArcGIS, GRASS, among others)

So, the first step will be to install QGIS in your laptop. This software is available for Windows, Mac OS X, Linux and Android. In all platforms, the environment is quite similar.

Download QGIS

Note, that you need to install some required packages before installing QGIS. Everything it is very well explained at the readme file.

For the impatient

• If you would like to start exploring QGIS, have a look to this manual Nice Introduction

Installing R and R-Studio

You probably already had this, yet I would just like to make sure that it is your computer. During the last lecture, we will introduce two packages in R to analyze GIS.

Other resources

• Replication codes and problems sets will be distributed via Github. If you are not familiar with this control version environment, it may be a nice opportunity to start using it. Go install it and don’t hesitate to comment or improve the codes from the course.

Installing python packages ( in case you )

Most of the available GIS solutions are python-based. Softwares such as ArcGIS and QGIS offer a wide range of pre-defined python syntaxis/functions to perform the most common task for GIS analysis. As we will discuss, these pre-defined functions face some restrictions that may impose an important loss of performance in your analysis. In particular, when you start running complex analysis that require more calculation power (e.g. multiprocessing, cluster based computer, among others).

This is why we should be able to program a standalone solutions that allow to integrate GIS tools for a more complex analysis in high-perform environments. We will use the following python packages:

* Pandas (A basic knowledge of this package will be assumed)
* GDAL
* Fiona
* rtree
* shapely (it is usually pre-installed)
* matplotlib

Installing packages for Python

• If you are windows-users, I suggest to install the conda. It is a package and environment manager for Python (and others languages like C).
  • The complete version is called Anacoda Anaconda
  • A small version is called Miniconda. One installed, you can install everyhing using the basic command Once you have installed you can install any python package just typing this

    conda install [package name]
    

    Or using the graphical environment provided by Anaconda

• For mac-users, I suggest to work with the pre-installed python distributor. So, don’t install anything. Just make sure you have the dependency manager called “pip”. Look here the instruction to install ot pip. As Anaconda, once install you can install any package using

conda install [package name]

• You can also install the packages manually. Just go to a python repository (e.g. this) and following the instruction from each package.