predictive discriminant analysis
a multivariate procedure that distinguishes between two or more categories of a future outcome on the basis of several current assessments of predictor or explanatory variables. More specifically, the technique combines variables measured at one point in time to produce the highest possible association with different levels or categories of an outcome measured at a subsequent point in time. For example, a researcher might measure daily levels of exercise, cholesterol, and average stress and use this information to predict which individuals will have a diagnosis of serious, mild, or no heart disease 6 months later. Compare descriptive discriminant analysis.