Discriminant analysis is applied to decide which variables differentiate between two or more naturally occurring groups. Discriminant analysis is a statistical analysis to forecast a definite dependent variable (called a grouping variable) by one or more continuous or binary independent variables (called predictor variables). The main objective of this type of analysis is to predict group membership dependent on a linear combination of the interval variables. The process starts with a set of observations where both group membership and the values of the interval variables are identified. The end result of the process is a model that permits extrapolation of group membership if the interval variables are known. This analysis also helps in understanding the data set, because thorough analysis of the prediction model that results because of this procedure is capable of giving an insight into the relationship between group membership and the variables used to predict group membership.
Mainly discriminant analysis is used to figure out the future buying behavior of a specific product by the customer or the likelihood of patient who might suffer from a specific disease.