A/B testing project on simulated data.
This was the task for the recruitment process. As I received great feedback, I decided to publish it to show my approach to the problem, although I had to erase some sensitive information. The analysis was performed on simulated data, so the outcomes are not realistic. That was my first A/B testing project and I know that there are were places for improvement.
Given the data from a control group, I had to provide a list of users which should be the most profitable during a new campaign.
I used R and its Tidyverse universe. Everything was written in a single RMarkdown notebook.
Distributed under the GNU GPLv3 license. See LICENSE
for more information.
I believe any contributions to this project do not make sense, however, feel free to use my code.