Overview
We now extend the concepts from Logistic Regression, where we describe how to build and use binary logistic regression models, to cases where the dependent variable can have more than two outcomes. Using such models the value of the categorical dependent variable can be predicted from the values of the independent variables.
In this part of the website, we focus on the case where there is order to these categories (ordinal logistic regression). E.g. the categories might be Strongly Disagree, Disagree, Agree, and Strongly Agree.
In Multinomial Regression, we turn our attention to the case where there is no order to the categories (multinomial logistic regression). E.g. the categories might be Christian, Muslim, and Jewish.
Topics
- Basic concepts and approaches for finding ordinal regression coefficients
- using binary logistic regression models
- general model using Solver
- using multinomial logistic regression
- proportional odds model using Solver
- Proportional odds model using Newton’s method
- Real Statistics capabilities