Logistic Regression

Also known as: Logit, Logistic Model, Binary Classification

Statistical model predicting the probability of a binary outcome (buy/don't buy, churn/no churn) based on predictor variables.

Logistic Regression is a classification statistical model that estimates the probability of a binary outcome (yes/no, buy/don't buy, churn/no churn) based on a set of predictor variables.

Unlike linear regression (which predicts continuous values), logistic regression transforms the linear combination of predictors through the sigmoid function to produce a probability between 0 and 1.

In market research, its applications include: predicting the probability of new product adoption from attitudes and demographic characteristics, identifying which factors best predict customer churn, and modeling purchase probability at different price points.

It is the conceptual foundation of discrete choice modeling and is also implemented within conjoint analysis models.