A form of Regression used when the Outcome Variable is a Categorical Variable. There are two qualitatively different forms and applications of this model:

- Modeling how preferences are determined by the features of alternatives (or, to use the language of Choice Modeling, to model how preferences are determined by the
*Attribute Levels* of the alternatives). For example, this model can be used to see how a phones' prices and features impact upon consumers' preferences for phones. This model is sometimes referred to as the *Conditional Logit* model.
- Modeling how categories in the Outcome Variable are related to characteristics of the observations in the sample. For example, this model can be used to model how choice of transport for commuting is determined by factors such as income, employment status, education and attitude to the environment. This model is sometimes referred to as
*Multinomial Logistic Regression* and *Multinomial Logistic Discriminant Analysis*.

## Next

Introduction to the Multinomial Logit Model

## Comments

0 comments

Article is closed for comments.