Multinomial regression is an extension of logistic regression that is used when a categorical outcome variable has more than two values and predictor variables are continuous or categorical. We can use multinomial regression to predict which of two or more categories a person is likely to belong to, compared to a baseline (or reference) category and given certain other information. With our longitudinal data we can use multinomial logistic regression to test the probability of an event occurring (A) in later life compared to other potential outcomes (B, C), applying information gathered in earlier life. In order to make comparisons, we can use any of the events (A, B or C) as the baseline category.
The Learning Hub is a resource for students and educators
tel | +44 (0)20 7331 5102 |
---|---|
closer@ucl.ac.uk |
Sign up for our email newsletters to get the latest from CLOSER
Sign up