Multiple Linear Regression
Content for Wednesday, October 25, 2023
Learning Goals
- Distinguish between confounders, colliders, and mediators
- Construct directed acyclic graphs (DAGs) using
dagitty.net
- Use DAGs to identify variables that need to be controlled for in regression
Week Overview
DATE | TIME | EVENT | MATERIAL |
---|---|---|---|
Mon, Oct 23 | 15:30-17:00 |
Consultation (Viktoriia) |
|
Tue, Oct 24 | 22:00 |
Problem Set 6 due |
|
Wed, Oct 25 | 13:45-15:15 |
Session: Multiple Linear Regression |
|
18:00 |
Problem Set 7 distributed |
||
Fri, Oct 27 | 22:00 |
Solutions Problem Set 6 available |
|
Mon, Oct 30 | 15:30-17:00 |
Consultation (Viktoriia) |
Slides
Reading Material
If you want to learn more about the topic, you are welcome to consult either The Effect book/videos or material by Andrew Heiss: - Chapter 6 in The Effect (Huntington-Klein 2021) - Chapter 7 in The Effect (Huntington-Klein 2021) - Chapter 8 in The Effect (Huntington-Klein 2021) - Chapter 9 in The Effect (Huntington-Klein 2021) - Youtube Playlist The Effect (starting from Causal Diagrams: Causality to Front Door Examples) - Chapter 10 Causal Inference in R for Political Data Science: A Practical Guide (Heiss 2021): This contains many of the examples from the lecture, like the campaign DAG (and more advanced stuff on alternatives to using statistical control in regression, such as matching) - More of the material from Andrew Heiss’ courses on the topic are available here: - https://evalf21.classes.andrewheiss.com/content/04-content/ - https://evalf21.classes.andrewheiss.com/example/dags/ - Statistical Rethinking 2023 - 06 - Good & Bad Controls by Richard McElreath - Bias mitigation in empirical peace and conflict studies: A short primer on posttreatment variables. (Dworschak): for some examples from conflict research paper
R for Data Analysis - Multiple Linear Regression R for Data Analysis - Multiple Linear Regression R for Data Analysis - Multiple Linear Regression R for Data Analysis