Sooner or later, every traffic analysis arrives at the question that really matters: did this visit turn into a customer, yes or no? We don’t care how much a session converted, we care whether it did. And right behind it comes the next question: which sessions are more likely to convert? The ones arriving from a search for the brand name? The ones browsing more pages? The ones on desktop?
With linear regression we learned to predict a number. Here, though, the outcome is a switch, on or off, and a straight line is not the right tool: we’ll see it even ends up predicting negative probabilities. The right tool is logistic regression, one of the most widely used models of all, from marketing to medicine. Let’s build it together, starting from an example with conversions.
Continue reading “Logistic Regression: What It Is, How to Read Odds Ratios, and an Example in R”