Multinomial Logistic Regression
- Used when the outcome has more than two categorical outcomes.
- Predicts the probability of membership in each category from predictor variables.
- Allows comparison of the relative importance of predictor variables.
Definition
Section titled “Definition”Multinomial logistic regression is a type of regression analysis used when there are multiple dependent variables, each with more than two categories. It is a way to predict the probability of an individual belonging to a certain category based on certain predictor variables.
Explanation
Section titled “Explanation”Multinomial logistic regression models the probabilities that an observation falls into each of several discrete categories of a dependent variable, using one or more predictor variables. It provides predicted probabilities for category membership and enables comparison of the relative importance of each predictor variable in explaining the dependent variable.
Examples
Section titled “Examples”Student grades
Section titled “Student grades”A researcher may want to predict the likelihood of a student being in the “A” grade, “B” grade, or “C” grade based on their gender, prior test scores, and extracurricular activities. In this case, the dependent variable would be the grade (with three categories: A, B, and C) and the predictor variables would be gender, test scores, and extracurricular activities.
Shampoo brand choice
Section titled “Shampoo brand choice”Predicting the likelihood of a customer choosing a certain brand of shampoo (A, B, or C) based on their age, income, and hair type. In this case, the dependent variable would be the brand of shampoo chosen (with three categories: A, B, and C) and the predictor variables would be age, income, and hair type.
Use cases
Section titled “Use cases”- Commonly used in fields such as psychology, marketing, and economics.
Related terms
Section titled “Related terms”- Predictor variable
- Dependent variable
- Probability