ABTesting
This portfolio captures the work I completed for a course, A/B Testing Design & Implementation, at Carnegie Mellon University in Fall 2021. The work done here involves fundamentals of randomized control trials (a.k.a. A/B tests), namely what they achieve, how to design, implement and analyze their outcomes as well as their shortcomings and work arounds. The assignments and project completed below also leverage the tools that can be used to analyze data from observational studies where randomization cannot be implemented. The course uses examples with real-world datasets drawn from research performed at the Heinz College in entertainment and education. To view my course repository on GitHub, please click here.
Key Learnings
From the course, A/B Testing Design & Implementation, I garnered an understanding of causality and the appropriate language to ask and answer causal questions. In addition to the theoretical knowledge, I learnt how the intuition is developed for implementation of fundamental tools required to measure causal effects. The following concepts have been covered in the assignments below: Randomized Control Experiments; Time and Individual Fixed Effects; Instrumental Variables; Natural Experiments; Differences in Differences; Propensity Score Matching; Compliance in Experiments; Heterogeneous Effects; Interference in Networked Experiments.
Portfolio
Here are few of the assignments that I completed during the course of this class. Additional assignments can be furnished upon request.
Assignments
To view the data, code and response file for each, please click on the hyperlinks below.
I. Time Dummies, Fixed Effects & First-Differences
Final Project
Brief Description: Using data from a Qualtrics survey executed by our team in Pittsburgh, the paper explores the causal effect of including scooter rides as part of the university transportation fee on weekly scooter usage. The causal inference in this scenario is established due to the elimination of selection bias, which was achieved through random assignment of experimental subjects to the treatment and control group. Moreover, the analysis extends to explore heterogeneity in the average causal effect across groups within the sample using key barriers of distance, safety and income as moderators. The project consisted of the following parts: Experiment Design & Causal Question of Interest; Survey Design & Execution; Data & Analysis and Recommendations. To view details, please click on the hyperlinks below: