multivariate regression

Multivariate regression models the relationship between one or more predictors and multiple outcome variables considered together. For example, a model might use a product's features to estimate both its expected demand and delivery time. Modeling outcomes jointly can represent relationships among those outcomes and may be useful when they share predictors or their errors are related. The term is sometimes used loosely to mean regression with several predictors, but that is more precisely called multiple regression when there is a single outcome; multivariate regression specifically involves multiple dependent variables. The modeling approach and evaluation measures depend on the structure of the outcomes and the question being asked.