A/B testing allows you to compare two or more versions of a given service to see which one performs better.
This repository includes A/B testing implementations for the Classical, Sequential, and ML approaches. To raise brand awareness, we used data collected by an advertising company running an online ad for a client. To boost its market competitiveness, the advertising company offers an additional service that quantifies the increase in brand awareness caused by the ads it displays to its online users.
The main goal is to create a reliable hypothesis testing algorithm to determine whether the advertisements run by the advertising company resulted in a significant increase in brand awareness. We will look at Classical, Sequential, and Machine Learning approaches to A/B testing.
Python 3.5
git clone https://github.com/10X-groups/AB-testing.git
cd SmartAd-Performance-Analysis
pip install -r requirements.txt