Alias

A term in an analysis of variance, regression or generalized linear model is said to be aliased if there is no information available to estimate its effects (or parameters) once the earlier terms in the model have been fitted.