Methods of Empirical Research in Economics

This course is ongoing, and material will be added and modified throughout. Registered students can find grades and additional material on UBC Canvas. TeX code for all documents is available from a git repository. Please contact me if you notice any errors or want to suggest changes or additions.

This page is updated often. It is a good idea to refresh (F5) it each time you visit.


Lecture notes and slides

  1. Introduction: slides
  2. Probability and statistics
    1. Probability slides, slides with notes
    2. Statistics and inference slides, slides with notes
  3. Simple linear regression: slides , slides with notes
  4. Multiple linear regression:
    1. Mechanics and interpretation slides, with notes
    2. Finite sample statistical properties my slides slides from Hiro Kasahara
    3. Asymptotic properties: slides, with notes
    4. Qualitative variables and functional form: slides
  5. Heteroskedasticity and dependence: slides, with notes
  6. Difference in differences: slides
  7. Instrumental variables: slides,with notes
  8. Conducting a research project
    1. slides on Levitt (1997) and McCrary (2002), data from McCrary (2002)
    2. slides on Triyana (2016)
    3. From past year: slides on AMS
    4. From past year: slides on Fang and Gavazza (2011)
    5. From past year: slides on Lundqvist, Dahlberg, and Mörk (2014)
  9. Regression discontinuity design slides
  10. Overview of other topics slides
  11. Review slides


Some students have been using the message board on Canvas to ask questions about the assignments. That's great, and I encourage you to use it to get and provide help. I will try to check the message board regularly, and answer questions when needed.
  1. Complete all chapters of Try R. This should not be difficult, but it will take some time. This assignment will not be graded, but completing it should make subsequent assignments much easier.
    You may also want to install R on your own computer. You can download R by itself from CRAN. The builtin text-editor of R is fairly limited, so you probably want to install an IDE or a nicer text editor. Rstudio is a very popular IDE for R.
  2. Remaining problem sets are on UBC Canvas. The first is due January 16, 2018.