The course is organised in height three-hours sessions. Each session articulates a theoretical lecture and practical applications using the R language. After each class, you will be expected to prepare exercises that will be corrected during the next class. The sessions are organised as follows:
Session
|
Topic
|
Duration
|
1
|
Statistical learning
|
3 hours
|
2
|
Advanced R
|
3 hours
|
3
|
Splines and GAM
|
3 hours
|
4
|
Trees, bagging and boosting
|
3 hours
|
5
|
Neural networks
|
3 hours
|
6
|
Deep learning
|
3 hours
|
7
|
Spatial data
|
3 hours
|
8
|
Spatial analysis
|
3 hours
|
All lectures, exercises, solutions and resources will be made available on the Dropbox of the course. You should come to class with your computer with administrator rights, as well as a working internet connection. Make sure to install the latest stable release of R, available on the CRAN website for all platforms. In addition, I strongly recommend that you install an IDE / GUI such as RStudio, also available for all platforms on their website.
GRADING
Individual report (75%)
You will have to submit a short report (15 pages maximum) by the end of the semester, along with a script reproducing your results. You will investigate a research question of your choice using some of the techniques studied in class. You are free to choose a topic related to your masters’ thesis, and to include some of this work in your thesis. In any case, your report should articulate statistical learning techniques and / or spatial analysis, along with traditional econometrics. You may use Rmarkdown to weave together text, code and output. Grading is based on the following criteria:
Criteria
|
Description
|
Question
|
Research question of interest to economists
|
Background
|
Short review of the related literature
|
Data
|
Data collection and construction of the datasets
|
Model
|
Choice of the appropriate model and specification
|
Results
|
Interpretation and answer to the research question
|
Writing
|
Clear and concise. English, referencing, etc.
|
Presentation of a paper (25%)
You will give a short presentation of a research paper at the beginning of a session. Make sure to use a presentation support and to give your presentation in less than 10 minutes.