A/B tests are very commonly performed by data analysts and data scientists.This project was done as part of the Udacity data analyst course. This project contains code which is used to understand the results of an A/B test run by an e-commerce website. The goal is to decide if the company should implement the new page, keep the old page, or perhaps run the experiment longer to make their decision.
You need to install Jupyter to run this code. You can install Anaconda which contains Jupyter.
AB Testing is a kind of hypothesis test. Here we use do a hypothesis test using two methods: bootstrapping the sample and calculating probabilities and through using the python library statsmodel and its z-score and p-value functions and compare the results of the two. Both methods give compatible results.
To access the code just click on the Jupyter Notebook above:
Analyze_ab_test_results_notebook.ipynb