YoVDO

Regression Discontinuity Design and Instrumental Variables

Offered By: Codecademy

Tags

Causal Inference Courses Data Analysis Courses

Course Description

Overview

Mimic experiments with the data you already have and measure the effects of treatment even with incomplete data.
A goal of many analytics projects is to answer "how much did factor x affect the outcome?", otherwise known as the treatment effect. But it can be very hard to do that with real data because it's often incomplete. This course will introduce you to two techniques to fix that problem: Regression Discontinuity Design and Instrumental Variables.



### Take-Away Skills
In this course, you will learn how and when to apply Regression Discontinuity Design (RDD) and Instrumental Variables. You'll be able to make the most of whatever data you have by mimicking experiments. You'll learn how and why cutoff points are so valuable in real-world data and how to get around assumptions like conditional exchangeability (defined in the course) with estimation techniques.

Syllabus

  • Regression Discontinuity Design and Instrumental Variables: Learn about regression discontinuity design and instrumental variables for causal inference.
    • Lesson: Regression Discontinuity Design
    • Quiz: RDD and IV Quiz
    • Project: Effect of Emergency Weather Systems on Transit Times
    • Lesson: Instrumental Variables
    • Article: Next Steps

Taught by

Kenny Lin

Related Courses

Data Science in Real Life
Johns Hopkins University via Coursera
A Crash Course in Causality: Inferring Causal Effects from Observational Data
University of Pennsylvania via Coursera
Causal Diagrams: Draw Your Assumptions Before Your Conclusions
Harvard University via edX
Causal Inference
Columbia University via Coursera
Causal Inference 2
Columbia University via Coursera