Analyze-Experiment-Results (A/B Test)

The Dataset

A/B tests are very commonly performed by data analysts and data scientists to ascertain which of two options should be chosen.

For this project, the A/B test is for an e-commerce website. The company has developed a new web page in order to try and increase the number of users who "convert," meaning the number of users who decide to pay for the company's product. The goal of this project is to analyze the data to help the company understand if they should implement this new page, keep the old page, or perhaps run the experiment longer to make their decision.

References

https://docs.w3cub.com/statsmodels/generated/statsmodels.stats.proportion.proportions_ztest/

https://pandas.pydata.org/pandas-docs/stable/reference/api/pandas.DataFrame.join.html

ros2/ros2cli#316

https://rebeccaebarnes.github.io/2018/05/01/what-is-a-p-value

https://numpy.org/doc/stable/reference/random/generated/numpy.random.binomial.html

https://github.com/Alicelibinguo/Analyzing-Website-Landing-Page-A-B-Test-Results-/blob/master/Analyze_ab_test_results_notebook.py

https://www.kaggle.com/tigerli1997/a-b-test-on-e-commerce-website