Hands-On Data Science for Marketing

Statistical hypothesis testing

The ultimate goal of A/B testing of different marketing strategies is to find out which strategy is the most efficient and works the best among the others. As briefly discussed in an earlier section, a strategy having a higher response number does not necessarily mean that it outperforms the rest. We will discuss how we can use the t-test to evaluate the relative performances of different marketing strategies and see which strategy wins over the others with significance.

In Python, there are two approaches to computing the t-value and p-value in a t-test. We will demonstrate both approaches in this section, and it is up to you to decide which one works more conveniently for you. The two approaches to compute ...